Editorial – October 2011 – Three of the MyOcean long time series reanalysis products
Greengs all,
This month’s newsleer is devoted to three of the MyOcean long me series Reanalysis products: the In Situ temperature and salinity CORA reanalysis
(1990 to 2010), the reanalysis of the North Atlanc ocean biogeochemistry (1998-2007) and the Arcc Ocean sea-ice dri/ reanalysis (1992-
2010).
The first product described here is the In Situ temperature and salinity CORA reanalysis (1990 to 2010). A new version of the comprehensive and
qualified ocean in-situ dataset (the Coriolis dataset for Re-Analysis - CORA) is released for the period 1990 to 2010. This in-situ dataset of temperature
and salinity profiles, from different data types (Argo, GTS data, VOS ships, NODC historical data...) on the global scale, is meant to be used for
general oceanographic research purposes, for ocean model validaon, and also for inializaon or assimilaon of ocean models. This product is
available from the MyOcean web portal (hp://www.myocean.eu/).
The second product is the reanalysis of the North Atlanc ocean biogeochemistry (1998-2007). A system assimilang Ocean Colour SeaWiFS data
during the period 1998-2007 has been designed to construct a reanalysis of the North Atlanc ocean biogeochemistry based on a coupled physicalbiogeochemical
model at eddy-admi:ng resoluon. The aim of this study is, on the one hand to develop the skeleton of a pre-operaonal coupled
physical-biogeochemical system with real-me assimilave/forecasng capability, and on the other hand to operate this prototype system for producing
a biogeochemical reanalysis covering the 1998-2007 period. This product is not available from the MyOcean web portal yet.
The third reanalysis product is the 1992-2010 winter Arcc Ocean sea ice dri/ me series made at Ifremer/CERSAT from satellite measurements
which consists of several products: the Level 3 products from single sensors and the L4 products from the combinaon of sensors. They are available
at 3, 6 and 30 day-lag with a 62.5 km-grid size during winter. This dataset is available for oceanic and climate modelling as well as various scienfic
studies in the Arcc. The me series is ongoing and will connue for Arcc long term monitoring using the next MetOp/ASCAT operaonal
scaerometers, planned to be operated for the next 10 years. This product is available from the MyOcean web portal (hp://www.myocean.eu/).
The next January 2012 issue will be dedicated to various applicaons using the Mercator Ocean products.
We wish you a pleasant reading!
Greetings all,
Once a year in April, the Mercator Ocean Forecasting Center in Toulouse and the Coriolis Infrastructure in Brest publish a common newsletter. Papers are dedicated to observations only.
• The first paper introducing this issue is presenting the Coriolis 2014-2020 framework which was renewed in 2014 in order to go on integrating in-situ ocean observation infrastructure for operational oceanography and ocean/climate research.
• Next paper by Poffa et al. describes how some Argo floats are deployed by the sailing community, through ship-based non-governmental organization or trans-oceanic races. It allows Argo floats to be deployed in poorly sampled areas where there is no regular shipping. Sailors got also involved in oceanographic science activities. An example of float deployment is given in the case of the Barcelona World Race.
• Next paper by Pouliquen et al. describes the EURO-ARGO ERIC infrastructure which is now officially set-up since May 2014. The objective of the Euro-Argo ERIC is to organize a long term European contribution to the international Argo array of profiling floats.
• Le Traon et al. are then presenting how the assessment of the impact of ARGO in Ocean models and satellite validation is conducted in the context of E-AIMS (Euro-Argo improvements for the GMES/Copernicus Marine Service) FP7 project. Observing System Evaluations and Observing System Simulation Experiments have been conducted to quantify the contribution of Argo to constrain global and regional monitoring and forecasting centers and validate satellite observations. Recommendations for the new phase of Argo are also elaborated.
• Kolodziejczyk et al. follow with the presentation of the complementarity of ARGO and SMOS Sea Surface Salinity (SSS) observations to help monitoring SSS variability from basin to meso scale. Using a 4-year time-series of SMOS SSS data and the global Argo array of in situ measurements, a statistical approach and an optimal interpolation method are used to characterize biases and reduce noises. Results are promising and show strong complementarity between SMOS and Argo data.
• Herbert et al. then describe Shipboard Acoustic Doppler Current Profilers (SADCP) observations which are carried out in the Tropical Atlantic during yearly cruises in the framework of the PIRATA program. The present note displays the SADCP data processing methodology applied for 8 PIRATA cruises by using CASCADE software.
• Cravatte et al. follow with a paper presenting the new international TPOS2020 project (2014-2020). The project objective is to build a renewed, integrated, internationally-coordinated and sustainable observing system in the Tropical Pacific, meeting both the needs of climate research and operational forecasting systems and learning lessons from the great success-and finally partial collapse- of the TAO/TRITON array.
• Saout-Grit et al. next present an updated procedure for CTD-oxygen calibration along with new
Editorial - May 2014 - Special Issue jointly coordinated by Mercator Ocean and Coriolis
focusing on Ocean Observations
Greetings all,
Once a year and for the fi fth year in a raw, the Mercator Ocean Forecasting Center in Toulouse and the Coriolis Infrastructure in Brest publish a
common newsletter. Some papers are dedicated to observations only, when others display collaborations between the 2 aspects: Observations and
Modelling/Data assimilation.
The fi rst paper by Cabanes et al. introducing this issue is presenting a new methodology aiming at correcting Argo fl oat salinity measurements in
delayed time when Argo fl oats conductivity sensors are subject to drift and offset due to bio-fouling or other technical problems.
Then, Cravatte et al. are using the Argo arrays in order to compile Argo fl oats’ drifts and show that they are a very valuable tool allowing determining
the absolute velocity. They apply this to study zonal jets at 1000 meters depth in the Tropics.
In the next paper, Maes and O’Kane provide with some results indicating the impact of a sustained ocean observing Argo network on the ability to
resolve the seasonal cycle of salinity stratifi cation by contrasting periods pre- and post-Argo. They take into account the respective thermal and saline
dependencies in the Brunt-Väisälä frequency (N2) in order to isolate the specifi c role of the salinity stratifi cation in the layers above the main pycno-
cline.
Picheral et al. are telling us about the Tara Oceans voyage that took place on the schooner “Tara” from 2009 to 2013 and visited all oceans. The ship
was adapted for modern oceanography. Scientifi c instruments were mounted on a dedicated CTD frame and installed on an underway fl ow-through
system. Data were sent daily to Coriolis. Post cruise calibrations were performed leading to a high quality dataset.
Then, Roquet et al. demonstrate the importance of the contribution of hydrographic and biogeochemical data collected by Antarctic marine mammals,
and in particular elephant seals, equipped with a new generation of oceanographic tags, for the environmental monitoring of the Southern Ocean.
The last paper of the present issue is displaying the collaboration between the Ocean Observations and Ocean Modelling communities: Turpin et
al. perform several Observing System Experiments in order to assess the impact of Argo observations on the Mercator Océan global analysis and
forecasting system at ¼ degree resolution.
We wish you a pleasant reading,
Laurence Crosnier and Sylvie Pouliquen, Editors.
#50
Newsletter
QUARTERLY
The Tara Oceans voyage took place on the schooner “Tara” from 2009 to 2013 and visited all oceans to collect samples and data in order to study the relationships between ecosystem biodiversity and function and the physical-chemical oceanographic environ-
ment (water mass, transport) (cf Picheral et al. this issue).
Credits: Francois Aurat/Tara Expéditions; Marc Picheral/LOV
Greetings all,
Once a year in April, the Mercator Ocean Forecasting Center in Toulouse and the Coriolis Infrastructure in Brest publish a common newsletter. Papers are dedicated to observations only.
• The first paper introducing this issue is presenting the Coriolis 2014-2020 framework which was renewed in 2014 in order to go on integrating in-situ ocean observation infrastructure for operational oceanography and ocean/climate research.
• Next paper by Poffa et al. describes how some Argo floats are deployed by the sailing community, through ship-based non-governmental organization or trans-oceanic races. It allows Argo floats to be deployed in poorly sampled areas where there is no regular shipping. Sailors got also involved in oceanographic science activities. An example of float deployment is given in the case of the Barcelona World Race.
• Next paper by Pouliquen et al. describes the EURO-ARGO ERIC infrastructure which is now officially set-up since May 2014. The objective of the Euro-Argo ERIC is to organize a long term European contribution to the international Argo array of profiling floats.
• Le Traon et al. are then presenting how the assessment of the impact of ARGO in Ocean models and satellite validation is conducted in the context of E-AIMS (Euro-Argo improvements for the GMES/Copernicus Marine Service) FP7 project. Observing System Evaluations and Observing System Simulation Experiments have been conducted to quantify the contribution of Argo to constrain global and regional monitoring and forecasting centers and validate satellite observations. Recommendations for the new phase of Argo are also elaborated.
• Kolodziejczyk et al. follow with the presentation of the complementarity of ARGO and SMOS Sea Surface Salinity (SSS) observations to help monitoring SSS variability from basin to meso scale. Using a 4-year time-series of SMOS SSS data and the global Argo array of in situ measurements, a statistical approach and an optimal interpolation method are used to characterize biases and reduce noises. Results are promising and show strong complementarity between SMOS and Argo data.
• Herbert et al. then describe Shipboard Acoustic Doppler Current Profilers (SADCP) observations which are carried out in the Tropical Atlantic during yearly cruises in the framework of the PIRATA program. The present note displays the SADCP data processing methodology applied for 8 PIRATA cruises by using CASCADE software.
• Cravatte et al. follow with a paper presenting the new international TPOS2020 project (2014-2020). The project objective is to build a renewed, integrated, internationally-coordinated and sustainable observing system in the Tropical Pacific, meeting both the needs of climate research and operational forecasting systems and learning lessons from the great success-and finally partial collapse- of the TAO/TRITON array.
• Saout-Grit et al. next present an updated procedure for CTD-oxygen calibration along with new
Editorial - May 2014 - Special Issue jointly coordinated by Mercator Ocean and Coriolis
focusing on Ocean Observations
Greetings all,
Once a year and for the fi fth year in a raw, the Mercator Ocean Forecasting Center in Toulouse and the Coriolis Infrastructure in Brest publish a
common newsletter. Some papers are dedicated to observations only, when others display collaborations between the 2 aspects: Observations and
Modelling/Data assimilation.
The fi rst paper by Cabanes et al. introducing this issue is presenting a new methodology aiming at correcting Argo fl oat salinity measurements in
delayed time when Argo fl oats conductivity sensors are subject to drift and offset due to bio-fouling or other technical problems.
Then, Cravatte et al. are using the Argo arrays in order to compile Argo fl oats’ drifts and show that they are a very valuable tool allowing determining
the absolute velocity. They apply this to study zonal jets at 1000 meters depth in the Tropics.
In the next paper, Maes and O’Kane provide with some results indicating the impact of a sustained ocean observing Argo network on the ability to
resolve the seasonal cycle of salinity stratifi cation by contrasting periods pre- and post-Argo. They take into account the respective thermal and saline
dependencies in the Brunt-Väisälä frequency (N2) in order to isolate the specifi c role of the salinity stratifi cation in the layers above the main pycno-
cline.
Picheral et al. are telling us about the Tara Oceans voyage that took place on the schooner “Tara” from 2009 to 2013 and visited all oceans. The ship
was adapted for modern oceanography. Scientifi c instruments were mounted on a dedicated CTD frame and installed on an underway fl ow-through
system. Data were sent daily to Coriolis. Post cruise calibrations were performed leading to a high quality dataset.
Then, Roquet et al. demonstrate the importance of the contribution of hydrographic and biogeochemical data collected by Antarctic marine mammals,
and in particular elephant seals, equipped with a new generation of oceanographic tags, for the environmental monitoring of the Southern Ocean.
The last paper of the present issue is displaying the collaboration between the Ocean Observations and Ocean Modelling communities: Turpin et
al. perform several Observing System Experiments in order to assess the impact of Argo observations on the Mercator Océan global analysis and
forecasting system at ¼ degree resolution.
We wish you a pleasant reading,
Laurence Crosnier and Sylvie Pouliquen, Editors.
#50
Newsletter
QUARTERLY
The Tara Oceans voyage took place on the schooner “Tara” from 2009 to 2013 and visited all oceans to collect samples and data in order to study the relationships between ecosystem biodiversity and function and the physical-chemical oceanographic environ-
ment (water mass, transport) (cf Picheral et al. this issue).
Credits: Francois Aurat/Tara Expéditions; Marc Picheral/LOV
The Development of a Catchment Management Modelling System for the Googong Re...GavanThomas
A scenario assessment model to assist the end-user in determining priorities for a series of agreed management prescriptions that can be enacted through controls on existing landuse
This the presentation I gave for my thesis defense. It\'s entitled "Using bioclimatic envelope modelling to incorporate spatial and temporal dynamics of climate change into conservation planning".
Application of GIS and MODFLOW to Ground Water Hydrology- A ReviewIJERA Editor
Groundwater is one of the most valuable natural resources, which supports human health, economic
development and ecological diversity. Due to over exploitation, the ground water systems are affected and
require management to maintain the conditions of ground water resources within acceptable limits. With the
development of computers and advances in information technology, efficient techniques for water management
has evolved. The main intent of the paper is to present a comprehensive review on application of GIS
(Geographic Information System) followed by coupling with MODFLOW package for ground water
management and development. Two major areas are discussed stating GIS applications in ground water
hydrology. (i) GIS based subsurface flow and pollution modelling (ii) Selection of artificial recharge sites.
Although the use of these techniques in groundwater studies has rapidly increased since last decade the sucess
rate is very limited. Based on this review , it is concluded that integation of GIS and MODFLOW have great
potential to revolutionize the monitoring and management of vital ground water resources in the future.
Greetings all,
This month’s newsletter is devoted to ocean indices aiming at a better understanding of the state of the ocean climate. Ocean
climate indices can be linked to major patterns of climate variability and usually have a significant social impact. The estimation of
the ocean climate indices along with their uncertainty is thus crucial: It gives an indication of our ability to measure the ocean. It is
as well a useful tool for decision making. Ocean climate indices also provide an at-a-glance overview of the state of the ocean
climate, and a way to talk to a wider audience about the ocean observing system. Several groups of experts are now working on
various ocean indicators using ocean forecast models, satellite data and reanalysis models in observing system simulation
experiments, among which the OOPC, NOAA and MERSEA/Boss4Gmes communities for example:
http://ioc3.unesco.org/oopc/state_of_the_ocean/index.php
http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/enso_advisory/
http://www.aoml.noaa.gov/phod/cyclone/data/method.html
http://www.mersea.eu.org/Indicators-with-B4G.html
Scientific articles about Ocean indices in the present Newsletter are displayed as follows: The first article by Von Schuckmann et
al. is dealing with the estimation of global ocean indicators from a gridded hydrographic field. Then, Crosnier et al. are describing
the need to conduct intercomparison of model analyses and forecast in order for experts to provide a reliable scientific expertise
on ocean climate indicators. The next article by Coppini et al. is telling us about ocean indices computed from the Mediterranean
Forecasting System for the European Environment Agency and Boss4Gmes. Then Buarque et al. are revisiting the Tropical
Cyclone Heat Potential Index in order to better represent the ocean heat content that interacts with Hurricane. The last article by Greiner et al. is dealing with the assessment of robust ocean indicators and gives an example with oceanic predictors for the
Sahel precipitations.
The next July 2009 newsletter will review the current work on data assimilation and its techniques and progress for operational
oceanography.
We wish you a pleasant reading.
Greetings to all,
The Global Ocean Data Assimilation Experiment (GODAE) final symposium will be held in Nice in November 12-15 2008. This
project has been a precursor to a world wide experiment to demonstrate the feasibility of global ocean observing systems using
state of the art assimilation techniques. Today, several teams are working on operational ocean systems to provide forecast and
description of the ocean, using increasingly complex assimilation schemes and high resolution models. As we saw in the last
newsletter, these systems have reached the coast and routinely provide real time ocean forecast. But they need input information
for their boundaries and initialisation fields, from regional, basin wide or global configurations.
This month, the Newsletter is dedicated to global ocean systems resulting from the GODAE project.
In the first news feature, a review of the GODAE achievements in ocean observing systems is made by Le Traon et al. In a
second introduction paper, Pierre Bahurel provides a “Global view on MyOcean” where he introduces the special ongoing efforts
to improve products and services to users.
Four systems from three countries (U.S., France and Japan) are then presented, showing a variety of developments, model
resolutions and assimilation schemes that are all facing the same challenges: to describe, understand and forecast the world
ocean. The first contribution is from Chassignet et Hurlburt and is dedicated to the U.S. HYCOM 1/12° global configuration.
Menemenlis et al. will then tell us how useful the ECCO2 system is in understanding and estimating ocean processes.
Legalloudec et al. follow with the 1/12° Mercator g lobal model and its ability to represent the mesoscale activity. Finally, Kamachi
et al. will present the MRI global systems, including two nesting configurations dedicated to several applications from climate
variability to boundary forcing or ocean weather.
The next newsletter will be published in January 2009 and dedicated to the Mediterranean Sea.
We wish you a pleasant reading.
Editorial – Apr 2013 – Special Issue jointly coordinated by Mercator Ocean
and Coriolis focusing on Ocean Observations
Greeti ngs all,
Once a year in April, the Mercator Ocean Forecasti ng Center in Toulouse and the Coriolis Infrastructure in Brest publish a common newslett er.
Papers are dedicated to observati ons only.
The fi rst paper introducing this issue is presenti ng the MyOcean InSitu Themati c Assembly Center (TAC) which is collecti ng and carrying out quality
control on In Situ data in a homogeneous manner and provides access to In Situ observati ons of core parameters in order to characterize ocean state
and its variability, thus contributi ng to initi alizati on, forcing, assimilati on and validati on of ocean numerical models.
Next paper by Kokoszka is using the MyOcean Observati on InSitu Themati c Assembly Center (TAC) Products in order to illustrate a strong wind event
in the North East Atlanti c in February 2013. Wind bursts over 100 km/h occurred along the French Atlanti c Coast on February 6th. A possible sea
surface temperature cooling is illustrated.
Next paper by Lebreton is dealing with the French Argo fl oat deployment from opportunity vessels. In 2012, Coriolis has deployed more than 120
fl oats using sailing, military or educati onal vessels. Deploying from opportunity vessels requires developing clear deployment procedures for teams
not familiar with Argo fl oats, training such teams to detect any anomaly in the deployment and parti cipati ng to outreach acti viti es.
Turpin et al. are then presenti ng MOOSE: A Mediterranean Ocean Observing System on Environment that has been set up as an interacti ve, distributed
and integrated observatory system of the North West Mediterranean Sea in order to detect and identi fy long-term environmental anomalies. It will
provide data for the MISTRALS project and use the MyOcean data distributi on infrastructure from the InSitu Themati c Assembly Center (TAC).
Reverdin et al. follow with the presentati on of SPURS, an experiment dedicated at improving our understanding of the processes controlling surface
salinity in the region of maximum surface salinity of the North Atlanti c subtropical region and at seeking how well remote sensing data can contribute in
monitoring and unraveling those processes. This includes a research cruise, STRASSE on board the French RV Thalassa, with a variety of measurements
of the upper ocean transmitt ed in real ti me, and a contributi on to the overall observing arrays of 10 surface drift ers, of 7 Argo fl oats, the use of two
merchant vessels crossing the area equipped with thermosalinographs and occasionally collecti ng XBT profi les.
Finally, Leymarie et al. are presenti ng the new ProvBioII fl oat with extended capaciti es among which a double board architecture as well as additi onal
batt ery and connectors. Collaborati on between LOV, NKE and Ifremer, and the opportunity off ered by the remOcean and NAOS projects, lead t
The Development of a Catchment Management Modelling System for the Googong Re...GavanThomas
A scenario assessment model to assist the end-user in determining priorities for a series of agreed management prescriptions that can be enacted through controls on existing landuse
This the presentation I gave for my thesis defense. It\'s entitled "Using bioclimatic envelope modelling to incorporate spatial and temporal dynamics of climate change into conservation planning".
Application of GIS and MODFLOW to Ground Water Hydrology- A ReviewIJERA Editor
Groundwater is one of the most valuable natural resources, which supports human health, economic
development and ecological diversity. Due to over exploitation, the ground water systems are affected and
require management to maintain the conditions of ground water resources within acceptable limits. With the
development of computers and advances in information technology, efficient techniques for water management
has evolved. The main intent of the paper is to present a comprehensive review on application of GIS
(Geographic Information System) followed by coupling with MODFLOW package for ground water
management and development. Two major areas are discussed stating GIS applications in ground water
hydrology. (i) GIS based subsurface flow and pollution modelling (ii) Selection of artificial recharge sites.
Although the use of these techniques in groundwater studies has rapidly increased since last decade the sucess
rate is very limited. Based on this review , it is concluded that integation of GIS and MODFLOW have great
potential to revolutionize the monitoring and management of vital ground water resources in the future.
Greetings all,
This month’s newsletter is devoted to ocean indices aiming at a better understanding of the state of the ocean climate. Ocean
climate indices can be linked to major patterns of climate variability and usually have a significant social impact. The estimation of
the ocean climate indices along with their uncertainty is thus crucial: It gives an indication of our ability to measure the ocean. It is
as well a useful tool for decision making. Ocean climate indices also provide an at-a-glance overview of the state of the ocean
climate, and a way to talk to a wider audience about the ocean observing system. Several groups of experts are now working on
various ocean indicators using ocean forecast models, satellite data and reanalysis models in observing system simulation
experiments, among which the OOPC, NOAA and MERSEA/Boss4Gmes communities for example:
http://ioc3.unesco.org/oopc/state_of_the_ocean/index.php
http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/enso_advisory/
http://www.aoml.noaa.gov/phod/cyclone/data/method.html
http://www.mersea.eu.org/Indicators-with-B4G.html
Scientific articles about Ocean indices in the present Newsletter are displayed as follows: The first article by Von Schuckmann et
al. is dealing with the estimation of global ocean indicators from a gridded hydrographic field. Then, Crosnier et al. are describing
the need to conduct intercomparison of model analyses and forecast in order for experts to provide a reliable scientific expertise
on ocean climate indicators. The next article by Coppini et al. is telling us about ocean indices computed from the Mediterranean
Forecasting System for the European Environment Agency and Boss4Gmes. Then Buarque et al. are revisiting the Tropical
Cyclone Heat Potential Index in order to better represent the ocean heat content that interacts with Hurricane. The last article by Greiner et al. is dealing with the assessment of robust ocean indicators and gives an example with oceanic predictors for the
Sahel precipitations.
The next July 2009 newsletter will review the current work on data assimilation and its techniques and progress for operational
oceanography.
We wish you a pleasant reading.
Greetings to all,
The Global Ocean Data Assimilation Experiment (GODAE) final symposium will be held in Nice in November 12-15 2008. This
project has been a precursor to a world wide experiment to demonstrate the feasibility of global ocean observing systems using
state of the art assimilation techniques. Today, several teams are working on operational ocean systems to provide forecast and
description of the ocean, using increasingly complex assimilation schemes and high resolution models. As we saw in the last
newsletter, these systems have reached the coast and routinely provide real time ocean forecast. But they need input information
for their boundaries and initialisation fields, from regional, basin wide or global configurations.
This month, the Newsletter is dedicated to global ocean systems resulting from the GODAE project.
In the first news feature, a review of the GODAE achievements in ocean observing systems is made by Le Traon et al. In a
second introduction paper, Pierre Bahurel provides a “Global view on MyOcean” where he introduces the special ongoing efforts
to improve products and services to users.
Four systems from three countries (U.S., France and Japan) are then presented, showing a variety of developments, model
resolutions and assimilation schemes that are all facing the same challenges: to describe, understand and forecast the world
ocean. The first contribution is from Chassignet et Hurlburt and is dedicated to the U.S. HYCOM 1/12° global configuration.
Menemenlis et al. will then tell us how useful the ECCO2 system is in understanding and estimating ocean processes.
Legalloudec et al. follow with the 1/12° Mercator g lobal model and its ability to represent the mesoscale activity. Finally, Kamachi
et al. will present the MRI global systems, including two nesting configurations dedicated to several applications from climate
variability to boundary forcing or ocean weather.
The next newsletter will be published in January 2009 and dedicated to the Mediterranean Sea.
We wish you a pleasant reading.
Editorial – Apr 2013 – Special Issue jointly coordinated by Mercator Ocean
and Coriolis focusing on Ocean Observations
Greeti ngs all,
Once a year in April, the Mercator Ocean Forecasti ng Center in Toulouse and the Coriolis Infrastructure in Brest publish a common newslett er.
Papers are dedicated to observati ons only.
The fi rst paper introducing this issue is presenti ng the MyOcean InSitu Themati c Assembly Center (TAC) which is collecti ng and carrying out quality
control on In Situ data in a homogeneous manner and provides access to In Situ observati ons of core parameters in order to characterize ocean state
and its variability, thus contributi ng to initi alizati on, forcing, assimilati on and validati on of ocean numerical models.
Next paper by Kokoszka is using the MyOcean Observati on InSitu Themati c Assembly Center (TAC) Products in order to illustrate a strong wind event
in the North East Atlanti c in February 2013. Wind bursts over 100 km/h occurred along the French Atlanti c Coast on February 6th. A possible sea
surface temperature cooling is illustrated.
Next paper by Lebreton is dealing with the French Argo fl oat deployment from opportunity vessels. In 2012, Coriolis has deployed more than 120
fl oats using sailing, military or educati onal vessels. Deploying from opportunity vessels requires developing clear deployment procedures for teams
not familiar with Argo fl oats, training such teams to detect any anomaly in the deployment and parti cipati ng to outreach acti viti es.
Turpin et al. are then presenti ng MOOSE: A Mediterranean Ocean Observing System on Environment that has been set up as an interacti ve, distributed
and integrated observatory system of the North West Mediterranean Sea in order to detect and identi fy long-term environmental anomalies. It will
provide data for the MISTRALS project and use the MyOcean data distributi on infrastructure from the InSitu Themati c Assembly Center (TAC).
Reverdin et al. follow with the presentati on of SPURS, an experiment dedicated at improving our understanding of the processes controlling surface
salinity in the region of maximum surface salinity of the North Atlanti c subtropical region and at seeking how well remote sensing data can contribute in
monitoring and unraveling those processes. This includes a research cruise, STRASSE on board the French RV Thalassa, with a variety of measurements
of the upper ocean transmitt ed in real ti me, and a contributi on to the overall observing arrays of 10 surface drift ers, of 7 Argo fl oats, the use of two
merchant vessels crossing the area equipped with thermosalinographs and occasionally collecti ng XBT profi les.
Finally, Leymarie et al. are presenti ng the new ProvBioII fl oat with extended capaciti es among which a double board architecture as well as additi onal
batt ery and connectors. Collaborati on between LOV, NKE and Ifremer, and the opportunity off ered by the remOcean and NAOS projects, lead t
Greetings all,
This month’s newsletter is devoted to Data Assimilation and its techniques and progress for operational oceanography.
Gary Brassington is first introducing this newsletter with a paper telling us about the international summer school for “observing,
assimilating and forecasting the ocean” which will be held in Perth, Western Australia in 11-22 January 2010
(http://www.bom.gov.au/bluelink/summerschool/). The course curriculum will include topics covering the leading edge science in
ocean observing systems, as well as the latest methods and techniques for analysis, data assimilation and ocean modeling.
Scientific articles about Data Assimilation are then displayed as follows: The first article by Broquet et al. is dealing with Ocean
state and surface forcing correction using the ROMS-IS4DVAR Data Assimilation System. Then, Cosme et al. are describing the
SEEK smoother as a Data Assimilation scheme for oceanic reanalyses. The next article by Brankart et al. is displaying a synthetic
literature review on the following subject: Is there a simple way of controlling the forcing function of the Ocean? Then Ferry et al.
are telling us about Ocean-Atmosphere flux correction by Ocean Data Assimilation. The last article by Oke et al. is dealing with
Data Assimilation in the Australian BlueLink System.
The next October 2009 newsletter will review the current work on ocean biology and biogeochemistry.
We wish you a pleasant reading!
Editorial – April 2011 – Special Issue jointly coordinated by Mercator Ocean and Coriolis
focusing on Ocean Observations
Greetings all,
Once a year in April, and for the second time after the April 2010 issue, the Mercator Ocean Forecasting Center in Toulouse and the Coriolis
Infrastructure in Brest publish a common newsletter. Some papers are dedicated to observations only, when others display collaborations
between the 2 aspects: Observations and Modelling/Data assimilation.
The two first papers introducing this issue are presenting the data requirement for the GMES Marine Core Service (Le Traon and Pouliquen) and
the Eurosites Open Ocean Observatory Network (Larkin et al.).
Then, Doxaran et al. are writing about the Provpanache project which uses of ProvBio floats to study the dynamics of suspended particles in river
plumes. Two papers are then dealing with eXpendable BathyThermograph (XBT) observations: Hamon et al. start with “Empirical correction of
XBT fall rate” and shows that maximum heat content in the top 700 meters found in earlier studies can be explained by now identified XBT
biases. XBT are also used by Maes et al. who look at the geostrophic component of oceanic jets entering in the eastern Coral Seas. Next, Brion
et al. are using complementary in situ data among which Thermosalinographs (TSG) for the calibration and validation of SMOS.
The two last papers of the present issue are displaying the collaboration between the Ocean Observations and Ocean Modelling communities:
Juza et al. are using a numerical model in order to determine how the Argo array could be extended to better monitor the Global Ocean heat
content variability. Drevillon et al. are then presenting the Mercator Ocean quaterly validation bulletin “Quo Va Dis?” which is using the Coriolis
data in order to draw the picture of the quality of the Mercator Ocean products.
We will meet again next year in April 2012 for a new jointly coordinated Newsletter between Mercator Ocean and Coriolis. Regarding next July
2011 Newsletter coordinated by Mercator Ocean only, it will display papers about the latest space missions and their use for oceanography and
research.
We wish you a pleasant reading,
Laurence Crosnier and Sylvie Pouliquen, Editors.
Summary:
ICE ARC Project
ASSESSING CLIMATE CHANGE IMPACTS ON MARINE ECOSYSTEMS AND HUMAN ACTIVITIES
IN THE ARCTIC OCEAN: THE EUROPEAN ACCESS PROGRAMME (2011-2015)
THE YEAR OF POLAR PREDICTION (YOPP): CHALLENGES AND OPPORTUNITIES
IN ICE-OCEAN FORECASTING
IAOOS (ICE - ATMOSPHERE - ARCTIC OCEAN OBSERVING SYSTEM, 2011-2019)
SEA ICE ANALYSIS AND FORECASTING WITH GLOSEA5
RECENT PROGRESS IN SEA ICE DATA ASSIMILATION AT ENVIRONMENT CANADA
RECENT DEVELOPMENTS IMPACTING THE SEA ICE IN THE MERCATOR OCÉAN GLOBAL ¼° CONFIGURATION
PARAMETERIZATION OF DRAG COEFFICIENTS OVER POLAR SEA ICE FOR CLIMATE MODELS
A MAXWELL-ELASTO-BRITTLE RHEOLOGY FOR SEA ICE MODELING
Greetings all,
This month’s newsletter is devoted to data assimilation and its application to Ocean Reanalyses.
Brasseur is introducing this newsletter telling us about the history of Ocean Reanalyses, the need for such Reanalyses for
MyOcean users in particular, and the perspective of Ocean Reanalyses coupled with biogeochemistry or regional systems for
example.
Scientific articles about Ocean Reanalyses activities are then displayed as follows: First, Cabanes et al. are presenting CORA, a
new comprehensive and qualified ocean in-situ dataset from 1990 to 2008, developped at the Coriolis Data Centre at IFREMER
and used to build Ocean Reanalyses. A more comprehensive article will be devoted to the CORA dataset in our next April 2010
issue. Then, Remy at Mercator in Toulouse considers large scale decadal Ocean Reanalysis to assess the improvement due to
the variational method data assimilation and show the sensitivity of the estimate to different parameters. She uses a light
configuration system allowing running several long term reanalysis. Third, Ferry et al. present the French Global Ocean
Reanalysis (GLORYS) project which aims at producing eddy resolving global Ocean Reanalyses with different streams spanning different time periods and using different technical choices. This is a collaboration between Mercator and French research
laboratories, and is a contribution to the European MyOcean project. Then, Masina et al. at CMCC in Italy are presenting the
implementation of data assimilation techniques into global ocean circulation model in order to investigate the role of the ocean on
climate variability and predictability. Fifth, Smith et al. are presenting the Ocean Reanalyses studies at ESSC in the U.K. which
aim at reconstructing water masses variability and ocean transports. Finally, Langlais et al. are giving an example of the various
uses of ocean reanalyses: they are using the Australian BlueLink Reanalysis in order to look into details at the various Southern
Ocean fronts.
The next April 2010 newsletter will introduce a new editorial line with a common newsletter between the Mercator
Ocean Forecasting Center in Toulouse and the Coriolis Data Center at Ifremer in Brest. Some papers will be
dedicated to observations only, when others will display collaborations between the 2 aspects: Observations and
Model. The idea is to wider and complete the subjects treated in our newsletter, as well as to trigger interactions
between observations and modeling communities. This common Mercator-Coriolis Newsletter is a test for which
we will take the opportunity to ask for your feedback.
We wish you a pleasant reading!
Greetings all,
This month’s newsletter is dedicated to Operational Oceanography around the world, with a focus on the European MERSEA
(http://www.mersea.eu.org/), the international GODAE (http://www.bom.gov.au/bmrc/ocean/GODAE/) programs as well as the
GMES (see Figure) (http://www.gmes.info/) initiative. Mercator Ocean is already fully involved in MERSEA as well as in GODAE
and GMES and will be one of the main actors of the Marine Core Service within the GMES project.
After an introduction by Desaubies reminding us of the challenges of the MERSEA project, this issue displays four articles giving
us a broad overview of what is done in the world of operational oceanography. We first start with an article illustrating how the
Mercator Ocean GMES Marine Core services are useful to downstream services such as Ocean climate monitoring, seasonal
prediction and support to offshore industry. We then follow with an article by our Australian colleagues (Brassington et al.)
involved in GODAE, describing their BlueLink operational oceanography system. Next article (Crosnier et al.) show how we
collaborate at the international GODAE level in order to inter-compare the forecasting systems, with an example of comparison of
the Mercator Ocean and BlueLink systems in the Indian Ocean. And last but not least article is written by our Canadian
colleagues (Davidson et al.) involved in the MERSEA and GODAE projects, describing their Canadian Ocean Forecasting
System. We wish you a pleasant reading and will meet again in April 2007, with a newsletter dedicated to impact studies.
Greetings all,
By the end of April 2008, the final meeting of the MERSEA European Project set up in Paris, in the Institut Océanographique.
The aim of the project was to develop a European system for operational monitoring and forecasting on global and regional scales
of the ocean physics, biogeochemistry and ecosystems.
It was surely a challenge to get together many different partners to build the future European operational oceanography of
tomorrow. It was also a challenge for the MERSEA teams to demonstrate their capacity to collect, validate and assimilate remote
sensed and in situ data into ocean circulation models, to interpolate in time and space for uniform coverage, to run nowcasting
(i.e. data synthesis in real-time), forecasting, and hind-casting, and to deliver information products. The project also had to
develop marine applications addressing the needs of both intermediate and end-users, whether institutional or from the private
sector
This Newsletter collects some of the many results obtained during this project. Several aspects are tackled: global and regional
forecasting systems, observations, and applications.
The News is written by the Coordinator of the Project, Yves Desaubies. He draws MERSEA results up.
In a first article, Marie Drévillon et al. present the MERSEA/Mercator-Ocean V2 global ocean analysis and forecasting system. In a
second one, Hervé Roquet et al. describe L3 and L4 high resolution SST products. The next article, written by Bruce Hackett et
al., focuses on Oil spill applications. The article of John Siddorn et al. closes the issue by a description of the development of a
North-East Atlantic tidal NEMO system.
Enjoy your reading!
Editorial – October 2010 – MyOcean Physical Systems in Regional Seas
Greetings all,
This month’s newsletter is devoted to some of the regional seas within the MyOcean project http://www.myocean.eu/ and to the numerical
systems that allows describing the ocean physics in those areas. A focus is here put on the Black Sea area, as well as on the Atlantic- Iberian
Biscay Irish- Ocean and the Atlantic- European North West Shelf- Ocean, with the description of new products that will be part of the MyOcean
catalogue from June 2011 on. Next January 2011 issue of the newsletter will also display papers about the Myocean project, focusing this time
on the ecosystem products in the Mediterranean Sea, the Arctic Ocean, the Black Sea as well as the Global Ocean.
In the present issue, Durand et al. are introducing this newsletter telling us about the Ferrybox data within MyOcean which are handled by the In-
Situ Thematic Assembly Centre (TAC). Environment sensor package onboard of ship-of-opportunity are referred to as Ferrybox system.
Ferrybox data products as temperature, salinity, oxygen and chlorophyll will be available from the MyOcean portal in near real time from June
2011. Delayed time products will be made available later on, around December 2011.
Scientific articles are then displayed as follows: First, Cailleau et al. are telling us about the numerical system that allows describing the ocean
physics in the MyOcean Atlantic- Iberian Biscay Irish- Ocean (referred to as the IBI area) from June 2011 on. Then, Demyshev et al. are talking
about the Black Sea MyOcean physics products used to investigate the Black Sea climatic changes. Results of the regional model improvement
and NEMO implementation in the Black Sea are also discussed. Finally, O’Dea et al. are presenting the next system for the Atlantic- European
North West Shelf- Ocean that will be part of the MyOcean catalogue from June 2011 on. A
The next January 2011 newsletter will also display papers about the Myocean project, focusing this time on the ecosystem products.
We wish you a pleasant reading!
C7.01: Current activities of the International Ocean Colour Coordinating Grou...Blue Planet Symposium
The International Ocean Colour Coordinating Group (IOCCG) was established in 1996 with the aim of developing consensus and synthesis on a global scale in the subject area of satellite ocean colour radiometry (OCR). It operates as an Affiliated Program of the Scientific Committee on Oceanic Research (SCOR) and comprises a rotating committee of representatives from each of the major international space agencies that provide ocean colour data, as well as representatives from the scientific community that use ocean colour data for research and applications. Space agencies contribute financially to the IOCCG and carry out the decisions endorsed by the group, while the scientific members address current research issues and make recommendations. Currently, IOCCG works towards ensuring Continuity and Consistency of the Ocean Colour Data Stream in the framework of the CEOS Ocean Colour virtual constellation. Within the OCR-VC framework, the International Network for Sensor Inter-comparison and Uncertainty Assessment for Ocean Color Radiometry (INSITU-OCR) initiative aims at integrating and rationalizing inter-agency efforts on satellite sensor inter-comparisons and uncertainty assessment for remote sensing products, with particular emphasis on requirements addressing the generation of ocean colour Essential Climate Variables (ECV) as proposed by the Global Climate Observing System (GCOS). Since 2013, IOCCG organises a bi-annual International Ocean Colour Science Meeting, where the global OCR community can gather and exchange with peers and space agency representatives. In parallel to these new initiatives, the IOCCG has a continuing capacity building and training activity, and continues to increase its record of monographs, based on the work of its working groups (currently 5 WG are active).
Greetings all,
This month’s newsletter is dedicated to regional and coastal oceanography. We review in this issue the impressive work
recently done towards regional to coastal modelling with nesting and open boundary procedures as well as imbrications of
models of increasing resolution and complexity. Moreover, regional and coastal systems have now reached an operational level
and are delivering real time forecast in various areas.
After an introduction by Obaton reminding us of the challenging European and French programs dealing with regional/coastal
oceanography, this issue displays six scientific articles. Chanut et al. are starting with a paper describing the Mercator Ocean
regional system embracing the French Atlantic coast with a 1/36° horizontal resolution. Marsaleix et a l. are then writing about
the North Western Mediterranean Sea system which is currently upgraded in the framework of the ECOOP program. Next paper
by Riflet et al. is dealing with operational ocean forecasting of the Portuguese waters using the Mercator Ocean North Atlantic
high resolution solution at its boundaries. Lecornu et al. are following with an article about the PREVIMER operational MARS
system in the Bay of Biscay. Marchesiello et al. are then describing the effort conducted at IRD in order to provide the
developing countries with tools for operational regional marine forecast. At last, Reffray et al. tell us how the MARS,
SYMPHONIE and NEMO/OPA systems intercompare over the Bay of Biscay during the year 2004.
We wish you a pleasant reading, and we will meet again in October 2008, with a newsletter dedicated to the international
GODAE project, which will hold its final meeting in Nice on November 12-15 2008 (http://www.godae.org/announcement-II.html).
Moreover, let us also remind you that our annual operational oceanography group meeting (Groupe Mission Mercator Coriolis,
GMMC) will take place on October 13 to 15 2008 in Toulouse (MétéoFrance site). We are looking forward to tell you about our
ongoing progress here at Mercator Ocean, and to hear about yours.
C1.02: The activities and Work Plan of the GOOS Physics and Climate Panel (OO...Blue Planet Symposium
The GOOS Physics and Climate Panel* coordinates requirements for the physical variables for GOOS, and also leads the climate theme, reporting to the United Nations Framework Convention for Climate Change (UNFCCC), through the Global Climate Observing System (GCOS).
Due to the many links that need to be made, the panel has a 5 year Work Plan (2013-2018), which is reviewed and updated annually. The Work Plan focusses on developing requirements for physics Essential Ocean Variables (EOVs) and ocean Essential Climate Variables (ECVs) through the development of variable specifications, the evaluation of network design to meet requirements, and the development and application of observing system performance metrics.
Where there is a need to evolve or expand the observing system, the panel conducts thematic systems based evaluations: for example, the Tropical Pacific observing system (TPOS) review, and evaluation of physics and climate aspects of the Deep Ocean Observing Strategy. Future plans and priorities include; Boundary Currents, and boundary current/shelf interactions, observations for reducing uncertainties in air sea fluxes, and observing ocean/ice interactions.
Systems based evaluations are designed to assess requirements for observations of EOVs, and assess approaches to meeting requirements with existing and emerging observing technologies. Some of these evaluations can and will lead to stand alone finite lifetime projects to oversee the transition or expansion of the observing system. For instance, a new project has been established following the TPOS review, called TPOS 2020 to oversee transition of the TPOS to become more robust, integrated and sustainable (see www.tpos2020.org).
* The GOOS Physics and Climate Panel is also known as the GCOS-GOOS-WCRP Ocean Observations panel for Physics and Climate, OOPC. See more details at www.oopc.info
Editorial – Jan/Feb/Mar 2013 – Impact of the loss/addition of satellite altimetry on operational
products
Greengs all,
This issue is dedicated to the study of the impact of the loss or addion of satellite almetry on operaonal products and systems.
The first news feature by Larnicol et al. is presenng the GODAE OceanView Observing System Evaluaon Task Team which primary objecve is to
support observaonal agencies by demonstrang the impact of observaons on operaonal forecast and reanalysis systems. Its secondary objecve
is to improve the performance of operaonal ocean forecast systems.
The second paper by Labroue et al. is reminding us about the main 2012 events within the satellite almetry constellaon. For the past two decades,
we have been used to take for granted the presence of several satellites flying together. The loss of Envisat in April 2012 and the decision to
put Jason-1 on its end of life orbit is a crude reminder of this constellaon fragility. Hence during 2012, the DUACS and MyOcean Sea Level TAC
teams have contributed to secure the almetry component in the frame of operaonal oceanography.
The third paper by Labroue et al. is displaying the potenal offered by Cryosat-2 for the mesoscale signal. The added value brought by Cryosat-2 as a
complement to the exisng almetry constellaon is discussed as well as how Cryosat-2 could contribute to secure the almetry constellaon and
thus the operaonal oceanography. Cryosat-2 mission has been introduced into the Near Real Time Sea Level system since February 2012 and has
been added to the Delayed Time system in April 2012.
The fourth paper by Remy et al. addresses the impact of the change of the satellite constellaon on the French Mercator Ocean analysis and forecas
ng systems. The impact of the loss of the ENVISAT and Jason1 along track Sea Level Anomaly data in the beginning of the year 2012 in the real
me products is studied. A dedicated set of Observing System Experiments (OSEs) is performed and preliminary results are shown. An OSE involves
running a copy of an exisng assimilaon run where some observaons are excluded. The difference between this run and the original run assimila
ng all the observaons allows a detailed assessment of the impact the observaons have on the assimilaon system.
Finally, the fi8h paper by Lea et al. is showing a number of Observing System Experiments (OSEs) to assess the impact of the observing network on
FOAM, the UK Met Office’s ocean assimilaon and forecasng system, as part of GODAE OceanView. A parallel version of the FOAM operaonal
system was run, during April 2011, withholding Jason-2 almeter observaons. Withholding Jason-2 removed 43% of the almeter data and resulted
in a 4% increase in the RMS SSH observaon-minus-background differences and around ±2ºC small scale changes in 100m temperature as well as
around ±0.2 psu changes in surface salinity.
We will meet again in April 2013 for a j
The 20th anniversary of the founding of Mercator
Océan (1995-2015) gives us an opportunity to
contemplate our past achievements but also to
look forward to the future. This issue has a special
meaning for all of us at Mercator Océan as it
pays tribute to men and women of the operational
oceanography community. We have thus portrayed
ten people you might not yet know, all of whom are
key actors (among many others) of today’s operational
oceanography and who are each worthy
of our attention.
We thus have an opportunity to thank all the scientists
who have published their work in the Mercator
Océan Newsletter and the Editorial Board*. We have
selected 24 papers to share with you again, sorted
into 10 themes. Through this issue, we intend to
highlight the work done over the last 20 years,
but above all to thank the people who did it, for
they are the actors who continually strive to build
today’s operational oceanography.
As you can imagine, selecting only 24 papers among
the last 53 issues was a tough choice for us!
This 20th anniversary also gives us an opportunity
to look ahead. The first Mercator Océan Newsletter
was published in April 2001. Fourteen years and
fifty-three issues later, it has become a reference
for a wide scientific community: each issue is read
by between 200 and approximately 5000 people
per year depending on the theme. To modernize
and streamline its circulation we have thus decided
to introduce the following changes.
Each issue will evolve with a more spacious and
easier to read page layout. The “Mercator Océan
Newsletter” is also changing its name and will
henceforth be called the “Mercator Ocean Journal”,
thus reflecting with a more appropriate term the
fact that it collates scientific papers. The editorial
line will not be changed, with 3 to 4 issues per year
publishing papers with a common theme as well
as an annual joint issue with the Coriolis Center
dedicated to in situ Observation. The first issue of
the “Mercator Ocean Journal” will focus on MyOcean2
and MyOcean Follow-on scientific output
and will be published in January 2016.
We sincerely hope you will enjoy this issue as much
as we have, for its content and the evocation of
all the work done over the past 20 years, but also
because it honors the dynamic and enthusiastic
scientists who each day add their contribution to
operational oceanography.
*Members of the Editorial Board are:
Bernard Barnier, CNRS, Directeur de Recherche, LGGE
Grenoble, France / Sylvie Pouliquen, Ifremer,
Head of Coriolis and EURO-ARGO ERIC Program Manager,
Brest, France / Pierre-Yves Le Traon, Scientific Director at
Mercator Océan, Toulouse, France / Gilles Garric, Innovation
Service Manager/R&D Dpt at Mercator Océan, Toulouse,
France / Laurence Crosnier, Product Manager
at Mercator Océan, Toulouse, France
UNDERSTANDING WHAT GREEN WASHING IS!.pdfJulietMogola
Many companies today use green washing to lure the public into thinking they are conserving the environment but in real sense they are doing more harm. There have been such several cases from very big companies here in Kenya and also globally. This ranges from various sectors from manufacturing and goes to consumer products. Educating people on greenwashing will enable people to make better choices based on their analysis and not on what they see on marketing sites.
Artificial Reefs by Kuddle Life Foundation - May 2024punit537210
Situated in Pondicherry, India, Kuddle Life Foundation is a charitable, non-profit and non-governmental organization (NGO) dedicated to improving the living standards of coastal communities and simultaneously placing a strong emphasis on the protection of marine ecosystems.
One of the key areas we work in is Artificial Reefs. This presentation captures our journey so far and our learnings. We hope you get as excited about marine conservation and artificial reefs as we are.
Please visit our website: https://kuddlelife.org
Our Instagram channel:
@kuddlelifefoundation
Our Linkedin Page:
https://www.linkedin.com/company/kuddlelifefoundation/
and write to us if you have any questions:
info@kuddlelife.org
Natural farming @ Dr. Siddhartha S. Jena.pptxsidjena70
A brief about organic farming/ Natural farming/ Zero budget natural farming/ Subash Palekar Natural farming which keeps us and environment safe and healthy. Next gen Agricultural practices of chemical free farming.
"Understanding the Carbon Cycle: Processes, Human Impacts, and Strategies for...MMariSelvam4
The carbon cycle is a critical component of Earth's environmental system, governing the movement and transformation of carbon through various reservoirs, including the atmosphere, oceans, soil, and living organisms. This complex cycle involves several key processes such as photosynthesis, respiration, decomposition, and carbon sequestration, each contributing to the regulation of carbon levels on the planet.
Human activities, particularly fossil fuel combustion and deforestation, have significantly altered the natural carbon cycle, leading to increased atmospheric carbon dioxide concentrations and driving climate change. Understanding the intricacies of the carbon cycle is essential for assessing the impacts of these changes and developing effective mitigation strategies.
By studying the carbon cycle, scientists can identify carbon sources and sinks, measure carbon fluxes, and predict future trends. This knowledge is crucial for crafting policies aimed at reducing carbon emissions, enhancing carbon storage, and promoting sustainable practices. The carbon cycle's interplay with climate systems, ecosystems, and human activities underscores its importance in maintaining a stable and healthy planet.
In-depth exploration of the carbon cycle reveals the delicate balance required to sustain life and the urgent need to address anthropogenic influences. Through research, education, and policy, we can work towards restoring equilibrium in the carbon cycle and ensuring a sustainable future for generations to come.
WRI’s brand new “Food Service Playbook for Promoting Sustainable Food Choices” gives food service operators the very latest strategies for creating dining environments that empower consumers to choose sustainable, plant-rich dishes. This research builds off our first guide for food service, now with industry experience and insights from nearly 350 academic trials.
DRAFT NRW Recreation Strategy - People and Nature thriving together
Mercator Ocean newsletter 43
1. #43
Editorial – October 2011 – Three of the MyOcean long time series reanalysis products
GreeƟngs all,
This month’s newsleƩer is devoted to three of the MyOcean long Ɵme series Reanalysis products: the In Situ temperature and salinity CORA reanal-
ysis (1990 to 2010), the reanalysis of the North AtlanƟc ocean biogeochemistry (1998-2007) and the ArcƟc Ocean sea-ice driŌ reanalysis (1992-
2010).
The first product described here is the In Situ temperature and salinity CORA reanalysis (1990 to 2010). A new version of the comprehensive and
qualified ocean in-situ dataset (the Coriolis dataset for Re-Analysis - CORA) is released for the period 1990 to 2010. This in-situ dataset of tempera-
ture and salinity profiles, from different data types (Argo, GTS data, VOS ships, NODC historical data...) on the global scale, is meant to be used for
general oceanographic research purposes, for ocean model validaƟon, and also for iniƟalizaƟon or assimilaƟon of ocean models. This product is
available from the MyOcean web portal (hƩp://www.myocean.eu/).
The second product is the reanalysis of the North AtlanƟc ocean biogeochemistry (1998-2007). A system assimilaƟng Ocean Colour SeaWiFS data
during the period 1998-2007 has been designed to construct a reanalysis of the North AtlanƟc ocean biogeochemistry based on a coupled physical-
biogeochemical model at eddy-admiƫng resoluƟon. The aim of this study is, on the one hand to develop the skeleton of a pre-operaƟonal coupled
physical-biogeochemical system with real-Ɵme assimilaƟve/forecasƟng capability, and on the other hand to operate this prototype system for pro-
ducing a biogeochemical reanalysis covering the 1998-2007 period. This product is not available from the MyOcean web portal yet.
The third reanalysis product is the 1992-2010 winter ArcƟc Ocean sea ice driŌ Ɵme series made at Ifremer/CERSAT from satellite measurements
which consists of several products: the Level 3 products from single sensors and the L4 products from the combinaƟon of sensors. They are available
at 3, 6 and 30 day-lag with a 62.5 km-grid size during winter. This dataset is available for oceanic and climate modelling as well as various scienƟfic
studies in the ArcƟc. The Ɵme series is ongoing and will conƟnue for ArcƟc long term monitoring using the next MetOp/ASCAT operaƟonal
scaƩerometers, planned to be operated for the next 10 years. This product is available from the MyOcean web portal (hƩp://www.myocean.eu/).
The next January 2012 issue will be dedicated to various applicaƟons using the Mercator Ocean products.
We wish you a pleasant reading!
Newsletter
QUARTERLY
The MyOcean Catalogue of products
hƩp://www.myocean.eu/web/24-catalogue.php
holds 215 PRODUCTS as of July 2011, among which many Reanalysis. This
catalogue will be updated on December 19 2011, with new and upgraded
products as well as new download services and a viewing capacity.
Credits: MyOcean hƩp://www.myocean.eu/
2. Quarterly Newsletter
Mercator Ocean
CONTENT
News: Workshop on Observing System Evaluation and Intercomparisons : a GODAE Ocean
view/GSOP/CLIVAR event (13-17 June 2011, Santa Cruz, USA)
CORA3, a comprehensive and qualified ocean in-situ dataset from 1990 to 2010
By Cécile Cabanes, Antoine Grouazel, Victor Turpin, François Paris, Christine Coatanoan, Karina Von Schuckmann,
Loic Petit de la Villeon, Thierry Carval, Sylvie Pouliquen
A multivariate reanalysis of the North Atlantic ocean biogeochemistry during 1998-2007
based on the assimilation of SeaWiFS data
By Clément Fontana, Pierre Brasseur, Jean-Michel Brankart
Ifremer/CERSAT Arctic sea ice drift 1992-2010 time series from satellite measurements for
myocean
By Fanny Girard-Ardhuin, Denis Croize-Fillon
3
4
10
20
3. #43—October 2011—3Mercator Ocean Quarterly Newsletter
NEWS: WORKSHOP ON OBSERVING SYSTEM EVALUATION AND INTER-
COMPARISONS : A GODAE OCEAN VIEW/GSOP/CLIVAR EVENT (13-17 JUNE
2011, SANTA CRUZ, USA)
Last June 2011, a workshop took place in Santa Cruz, USA, on the Observing System Evaluation and Intercomparisons. It was jointly organised
by the Climate Variability Global Synthesis and Observations Panel (M. Balmaseda, CLIVAR-GSOP) as well as two of the GODAE Ocean view
task teams: The Observing System Evaluation Task Team (P. Oke) and the Intercomparison and Validation Task Team (M. Martin). One of the
main objectives of the workshop was to demonstrate and assess the value of observations to ocean forecast and analysis systems. Workshop
report and presentations are available at the following address:
https://www.godae-oceanview.org/outreach/meetings-workshops/godae-oceanview-gsop-clivar-workshop/.
Specific goals of the workshop were to:
• Demonstrate the value of in-situ and satellite observations to short term, seasonal and decadal forecast systems.
• Move towards routine monitoring of the global ocean observing system (explore possibilities for “Observations Impact statements”).
• Review intercomparison of class 4 metrics from operational short-range forecast systems and seasonal prediction systems.
• Linking intercomparison and observing system evaluation monitoring activities within GODAE OceanView to those of CLIVAR.
Many studies were presented that demonstrate the value of the global ocean observing system for short term to seasonal forecasts. Overall, the
main outcomes of the workshop are:
⇒ The ocean forecasting and reanalysis community needs a common framework for reporting their results to the observational commu-
nity and agencies that are responsible for maintaining the GOOS (Global Ocean Observing System), despite the fact that the value of
different type of observations (like ARGO, altimetry, Sea Surface Temperature) is already demonstrated. There was an agreement to
find a way towards developing “Observing Impact Statements”.
⇒ Routine monitoring of the impact of ocean observations on the forecasting and analysis systems has been developed through a range
of methods, most of which are derived from data assimilation theory. A successful demonstration has been done by the UK-MetOffice.
Then several operational centres state their intention to participate to the near real time OSE (Observing System Experiments) activi-
ty.
⇒ The value of international intercomparison activities was highlighted by the great success of the CLIVAR-GSOP community. These
intercomparisons help to clarify the common problems in the systems, and to assess the impact of the observations. It was agreed
that the GODAE OceanView community should work towards the generation of a multi-model ensemble. In addition, there was an
agreement to produce and intercompare various climate metrics from GOV (Godae Ocean View) and GSOP (Global Synthesis and
Observations Panel).
News: Workshop on Observing System EvaluaƟon and Intercomparisons : a GODAE Ocean view/GSOP/CLIVAR event (13-17 June 2011, Santa Cruz, USA)
4. #43—October 2011—4Mercator Ocean Quarterly Newsletter
CORA3, A COMPREHENSIVE AND QUALIFIED OCEAN IN-SITU DATASET
FROM 1990 TO 2010
By Cécile Cabanes (1)
, Antoine Grouazel (1)
, Victor Turpin(2)
, François Paris(2)
, Christine Coatanoan (2)
, Karina
Von Schuckmann(1)
, Loic Petit de la Villeon (2)
, Thierry Carval (2)
, Sylvie Pouliquen (1)
1
Laboratory of Oceanography from Space (LOS), IFREMER/CNRS, Brest, France
2
Coriolis Data Centre, IFREMER, Brest, France
Abstract
Coriolis is a French program which aims at contributing to the ocean in situ measurements as part of the French oceanographic operational sys-
tem. It has been especially involved in gathering all global ocean in-situ observation data in real time, and developing continuous, automatic, and
permanent observation networks.
A new version of the comprehensive and qualified ocean in-situ dataset, the Coriolis dataset for Re-Analysis (CORA), is produced for the period
1990 to 2010. This in-situ dataset of temperature and salinity profiles, from different data types (Argo, GTS data, VOS ships, NODC historical
data...) on the global scale, is meant to be used for general oceanographic research purposes, for ocean model validation, and also for initializa-
tion or assimilation of ocean models.
To generate this new version, new and updated data have been extracted from the Coriolis database and added to the previous CORA dataset
spanning the period 1990-2008. To qualify this dataset, several tests have been developed to improve in a homogeneous way the quality of the
raw database and to fit the level required by the physical ocean re-analysis activities. These tests include some simple systematic tests, a test
against climatology and a more elaborate statistical test involving an objective analysis method (for the validation of this dataset). Visual quality
control (QC) is performed on all the suspicious temperature (T) and salinity (S) profiles and quality flags are modified in the dataset if necessary.
This Coriolis product is available on request through MyOcean Service Desk (http://www.myocean.eu/).
Introduction
The Coriolis database is a real time dataset as it is updated every day as new data arrive. On the contrary, the CORA database corresponds to an
extraction of all in situ temperature and salinity profiles from the Coriolis database at a given time. All the data are then re-qualified. CORA is meant to fit
the needs of both re-analysis and research projects. However, dealing with the quantity of data required by re-analysis projects and the quality of
data required by research projects, remains a difficult task.
Several important changes have been made since the last release of CORA2, concerning both the production procedure (to be able to release
yearly reanalysis) and quality checks applied to the data. Those changes are the following:
• A new procedure is now used to produce the dataset: This procedure was set up to be able to extract only new and updated data from the
Coriolis database at each new release of CORA.
• A new set of quality checks is performed on the data.
• A check of duplicates was re-run on the whole dataset
• An XBT bias correction has been applied.
• The CORA3 release has been extended for the period 1990-2010.
CORA3, a comprehensive and qualified ocean in-situ dataset from 1990 to 2010
5. #43—October 2011—5Mercator Ocean Quarterly Newsletter
Description of the Dataset
The Coriolis centre receives data from the Argo program, French research ships, GTS data, GTSPP, GOSUD, MEDS, voluntary observing and
merchants ships, moorings, and the World Ocean Database (not in real time for the last one). CORA thus contains data from different types of
instruments: mainly Argo floats, XBT, CTD and XCTD, and moorings. The data are stored in 7 files types: PF, XB, CT, OC, MO, BA, and TE. The
data from Argo floats directly received from DACS (PF files) have a nominal accuracy of 0.01°C and 0.0 1 PSU and are transmitted with full resolu-
tion. XBT or XCTD data received from research and opportunity vessels (XB files) provide an accuracy of 0.03°C to 0.1°C for temperature and
0.03 to 0.1 PSU for salinity. The CT files contain CTD data from research vessels (accuracy on the order of 0.001°C for temperature and 0.001
PSU for salinity after calibration) but also data from sea mammals equipped with CTDs (accuracy is on the order of 0.005°C for temperature and
0.02 PSU for salinity but can be lower depending of the availability of reference data for post-processing, see Boehme et al., 2009) and some Sea
Gliders. Other CTD data are found in OC files and come from the high resolution CTD dataset of the World ocean database 2009 (Boyer et al.,
2009). Mooring data (MO files) are mostly from TAO TRITON RAMA and PIRATA moorings and have accuracy comparable to Argo floats. The
two last categories (TE and BA files) are for all data transmitted trough the GTS (data from Argo floats not yet received at the DACS, moorings,
etc...). This transmission system imposes limitations on the accuracy: data are truncated respectively two and one places beyond decimal point
for TE and BA type. Figure 1 shows the number of temperature and salinity profiles in the CORA3 database for the whole period 1990-2010 and
their repartition among the different file types. A large amount of data comes from the GTS as a consequence of the real time needs of the Corio-
lis data centre.
Figure 1: A) Number of profiles in the Global
Ocean in the different file types in CORA3 and B)
Number of temperature and salinity data per
month over the global Ocean, as a funcƟon of
depth (from S.Guinehut, CLS, Toulouse).
CORA3, a comprehensive and qualified ocean in-situ dataset from 1990 to 2010
6. #43—October 2011—6Mercator Ocean Quarterly Newsletter
Description of data processing
Data flow
CORA data are retrieved from the Coriolis database which is constantly evolving because new data are submitted, some data are reprocessed,
other data are adjusted in delayed mode (Argo data), quality flags are modified in the Coriolis database after quality checks have been performed,
etc…. Basically, all new and modified data since the previous version of CORA are retrieved from the Coriolis database. This subset of new and
updated data are then re-qualified. Other quality checks are performed on the whole dataset to ensure data quality consistency
Quality checks
Data received by the Coriolis data centre from different sources are put through a set of quality control procedures (Coatanoan and Petit de la
Villéon, 2005) to ensure a consistent dataset.
Beside those tests, several other quality checks have been developed or applied to produce CORA3 in order to reach the quality level required by
the physical ocean re-analysis activities. These checks include some simple systematic tests; a test against climatology and a more elaborate
statistical test involving an objective analysis method (see Gaillard et al., 2009 for further details). Visual quality control (QC) is performed on all
suspicious temperature and salinity profiles. After these visual checks it is decided to change or not the control quality flag.
Systematic test on new and updated data
A profile fails a systematic test when pressure is negative, T and S values are outside an acceptable range depending on depth and region, T or S
are equal to zero at bottom or surface, values are constant at depth, values are outside the 10 s climatological range, if there is large salinity gra-
dient at the surface (more than 5 PSU within 2dB) or a systematic bias. Each time a profile failed a systematic test it was visually checked.
Tests on the whole database
A test against climatology that we call Anomaly Method was also applied. In this case, a profile failed the test if at least 50% of its data points lie
outside the 5 s climatological range. This allows detecting smaller deviations compare to the 10 s check. The statistical test is based on an objec-
tive analysis run (Bretherton et al., 1976) with a three weeks window. Residuals between the raw data and the gridded field are computed by the
analysis. Residuals larger than a defined value produce alerts that are then checked visually. This method combines the advantage of a colloca-
tion method since it takes into account all neighbouring sensors, and the comparison with climatology. Finally, Argo floats pointed out by the alti-
metric test (Guinehut et al., 2009 and ftp://ftp.ifremer.fr/ifremer/argo/etc/argo-ast9-item13- AltimeterComparison/) were systematically verified over
all their life period and quality control flags were modified when necessary.
Quality control for the CORA3 database also includes feedbacks from N. Ferry (Mercator Ocean, Toulouse) who performed the run GLORYS2V1.
This feedback provides a list of suspicious profiles detected by a comparison with the model solution. About 50% of the alerts where confirmed
after visual check for the CORA3 database. Figure 2 is an example of a suspicious profile detected. This kind of feedback can help to improve our
tests.
Figure 2: Example of a suspicious profile provid-
ed by the feedback of N. Ferry (Mercator
Ocean, Toulouse) who performed the GLO-
RY2V1 run. Salinity innovaƟon (PSU) (leŌ panel)
and the visual check performed on this alert
(right panels: temperature in °C and salinity in
PSU as a funcƟon of depth).
CORA3, a comprehensive and qualified ocean in-situ dataset from 1990 to 2010
7. #43—October 2011—7Mercator Ocean Quarterly Newsletter
Elimination of redundant profiles
A last test was performed on the whole CORA dataset in order to check if double data exist. This test follows different steps:
1.All data pairs following a space-time criterion are found (the space time criterion depends on the data type).
2.For all pairs it is checked if it is a double or not: temperature and salinity profile mean and standard deviation are computed after inter-
polation on same depth levels, if necessary. If the differences between the profile means and standard deviations are lower than the
threshold values then the pair is considered as a double.
3.If the pair is a double, then it is decided which station has to be removed from the database.
4.To make such a choice, the type of station is first considered (stations from the GTS - 'BA' or 'TE' types - are of lower precision), then
the number of parameters associated to the station is considered (For example, keep the station with TEMP and PSAL instead of the
station with only TEMP data) and finally the station that reach maximum depth is kept.
XBT bias correction
Different issues with the data of eXpendable BathyThermograph (XBTs) exist and, if not corrected, they are known to contribute to anomalous
global heat content variability (e.g Wijffels et al. 2008). The XBT system measures the time elapsed since the probe entered the water and thus
inaccuracies in the fall rate equation result in depth errors. There are also issues of temperature offset but usually with little dependence on depth.
The correction applied on CORA3 dataset is an application of the method described in Hamon et al., 2011. This correction is divided in two parts:
first the computation of the thermal offset then the correction of depth. To evaluate the temperature offset and the error in depth the reference
used are all the co-localised profiles (e.g. in a 3km ray, +/-15 days temporal frame, a maximum average temperature difference of 1°C and a
bathymetric difference inferior to 1000m) that are not XBT and with quality flags different from 3 and 4 (suspicious and bad quality). Those refer-
ences thus gather CTD, Argos profilers and mooring buoys.
The temperature offset correction:
This correction aims to give a value of correction for each profile as a function of the year and the category of XBT: shallow XBT or deep XBT (e.g.
maximal depth >=500m). The values are computed by the difference of each XBT profile with its reference profile in the layer 30-50m (below the mixed
layer and where depth errors are not important enough to parasite values of temperature). Solely low temperature gradient points (e.g. <0.0025°
C/m) are used to compute those corrections. XBT and reference profiles are interpolated on standard levels from 0 to 1000m and a resolution of
10m before the calculation. The finaloffsetisthemedianofallthosedifferences.
The depth errorcorrection:
Results of this depth correction are second order polynomial coefficients depending on the year, the depth and the category of XBT profiles. In this second step there
arenottwobutfourdifferentcategories:deepandhot,deepandcold,shallowandhot, shallowandcold(e.g.maximaldepth500mandmeantemperature11°C).Toevaluate
theerrorindepthweusethefollowingformula:
Somecursors such as aminimum number of depth errors perlevel and amaximum value of depth error permit to improve the quality of the median profile resulting from
the raw depth errors in each of the four categories andat eachlevel.Thenwefitasecondorderpolynomialonthemediandeptherrors andwegetthreecoefficientsa,band
c:
In the layer 30-200m, the coefficient is replaced by the mean of the depth error in order to ignore the noise due to the mixed layer.
refz
ref
T
TT
dZ
∂
−
=
cbZaZZZtrue
++=− 2
CORA3, a comprehensive and qualified ocean in-situ dataset from 1990 to 2010
8. #43—October 2011—8Mercator Ocean Quarterly Newsletter
Figure 3 shows the impact of the correction on the difference between XBT and reference profiles in the CORA3 database as a function of time
and depth. The large positive bias is largely reduced.
Examples: Use of the CORA data base
Research
Using the CORA database to estimate global ocean temperature, heat and freshwater content is a way to asses the quality of the dataset as the-
se global quantities are very sensitive to any sensor drift or systematic instrumental bias (see Von Schuckmann and Le Traon, 2011). Although
our quality controls are meant to detect such instrument problems, they can still miss small unknown drifts or bias. Comparison and sensitivity
studies are thus of primary importance in the domain of climatic changes. In the Von Schuckmann and Le Traon (2011) study, Argo temperature
and salinity measurements during the period 2005 to 2010 were used to estimate Global Ocean indicators (GOIs) such as global ocean heat con-
tent (GOHC), global ocean freshwater content (GOFC) and global steric sea level (GSSL). A method based on a simple box averaging scheme
using a weighted mean. Uncertainties due to data processing methods and choice of climatology are estimated. This method is easy to implement
and run and can be used to set up a routine monitoring of the global ocean. Over the six year time period, trends of GOHC and GSSL are 0.55 +/-
0.1 W/m2 and 0.69 +/- 0.14 mm/yr, respectively. The trend of GOFC is barely significant. Results show that there is significant interannual variabil-
ity at global scale, especially for GOFC. Annual mean GOIs from the todays Argo sampling can be derived with an accuracy of 0.1 cm for GSSL,
0.21x108
J/m2 for GOHC, and 700 km3 for GOFC. Long-term trends (15 years) of GOIs based on the complete Argo sampling for the upper
1500m depth can be estimated with an accuracy of 0.03 mm/yr for GSSL, 0.02 W/m2 for OHC and 20 km3/yr for GOFC - under the assumption
that no systematic errors remain in the observing system.
A method that combines vertical profiles from the CORA dataset and satellite Sea Surface Height (SSH) data was also used to reconstruct the
vertical structure of mesoscale eddies in the South Atlantic Ocean (see Azevedo Correia et al., 2011). This allowed estimating an annual mean
eddy meridional heat flux and a volume transport from the Indian to the Atlantic Ocean and describing its interannual variability.
Data assimilation in ocean models
An important application of such a database is also its use for ocean reanalyses. Throughout the world, several reanalyses projects are underway
which aim at providing a continuous space-time description of the ocean, synthesizing the information provided by various observation types
(remotely sensed and in situ) and the constraints provided by the physics of numerical ocean models. In France, global ocean reanalysis activity
is a joint collaboration between Mercator Océan, Coriolis data centre and several oceanographic and atmospheric research laboratories in the
framework of GLORYS (GLobal Ocean ReanalYsis and Simulations) project. This project contributes also to the production of coordinated reanal-
yses at the European level in the context of MyOcean EU funded FP7 project, in collaboration with Italian, English, French and Canadian part-
ners. The goal of GLORYS is to produce a series of realistic eddy resolving global ocean reanalyses. Several reanalyses are planned, with differ-
ent streams. Each stream can be produced several times with different technical and scientific choices. Stream 2 (GLORYS2) covering the period
1992-2010 has been produced using the CORA3 data set.
Figure 3: Temperature difference (in °C) as a
funcƟon of Ɵme (years) and depth (in meters)
between XBT and reference profiles before (A)
and aŌer (B) correcƟon.
CORA3, a comprehensive and qualified ocean in-situ dataset from 1990 to 2010
9. #43—October 2011—9Mercator Ocean Quarterly Newsletter
References
Azevedo Correia J.M., C. De Boyer Montegut, C. Cabanes, P. Klein : Estimation of the Agulhas ring impacts on meridional heat fluxes and
transport using ARGO floats and satellite data, submitted to GRL August 2011.
Boehme, L P. Lovell, M. Biuw, F. Roquet, J. Nicholson, S. E. Thorpe, M. P. Meredith, and M. Fedak, 2009: Technical Note: Animal-borne CTD-
Satellite Relay Data Loggers for real-time oceanographic data collection,Ocean Sci., 5, 685–695
Boyer, T. P., J. I. Antonov , O. K. Baranova, H. E. Garcia, D. R. Johnson, R. A. Locarnini, A. V. Mishonov, T. D. O’Brien, D. Seidov, I. V. Smolyar,
M. M. Zweng, 2009. World Ocean Database 2009. S. Levitus, Ed., NOAA Atlas NESDIS 66, U.S. Gov. Printing Office, Wash., D.C., 216 pp.,
DVDs.
Bretherton, F., R. Davis, and C. Fandry, 1976: A technique for objective analysis and design of oceanic experiments applied to Mode-73. Deep-
Sea Res., 23, 559–582.
Coatanoan C., and L. Petit de la Villéon, 2005: Coriolis Data Centre, In-situ data quality control procedures, Ifremer report, 17 pp. (http://
www.coriolis.eu.org/cdc/documents/cordo-rap-04-047-quality-control.pdf)
Gaillard F., E. Autret, V. Thierry, P. Galaup, C. Coatanoan, and T. Loubrieu, 2009: Quality controls of large Argo datasets, J. Atmos. Ocean.
Tech., 26, 337-351.
Guinehut, S., C. Coatanoan, A.-L. Dhomps, P.-Y. Le Traon and G. Larnicol, 2009: On the Use of Satellite Altimeter Data in Argo Quality Control,
Journal of Atmospheric and Oceanic Technology, 26, 395-402
Hamon M., G. Reverdin, P. Y. Le Traon : Empirical correction of XBT data, submitted to JGR June 2011
Von Schuckmann K. and P-Y. Le Traon, 2011: How well can we derive Global Ocean Indicators from Argo data?, Ocean Sci. Discuss., 8, 999–
1024.
Wijffels, Susan E., J. Willis, C. M. Domingues, P. Barker, N. J. White, A. Gronell, K. Ridgway, and J. A. Church, 2008: Changing Expendable
Bathythermograph Fall Rates and Their Impact on Estimates of Thermosteric Sea Level Rise. J. Climate, 21, 5657–5672. doi:
10.1175/2008JCLI2290.1
CORA3, a comprehensive and qualified ocean in-situ dataset from 1990 to 2010
10. #43—October 2011—10Mercator Ocean Quarterly Newsletter
A MULTIVARIATE REANALYSIS OF THE NORTH ATLANTIC OCEAN BIOGEO-
CHEMISTRY DURING 1998-2007 BASED ON THE ASSIMILATION OF SEAWIFS
DATA
By Clément Fontana(1)
, Pierre Brasseur(1)
, Jean-Michel Brankart(1)
(1)
LEGI-CNRS, Grenoble, France
Abstract
In this paper, we describe an assimilative system that has been designed to construct a reanalysis of the North Atlantic ocean biogeochemistry
based on satellite ocean colour data and a coupled physical-biogeochemical model at eddy-admitting resolution. Two distinct experiments, which
differ by the nature of the analysis scheme implemented in a simplified SEEK filter, have been assimilating SeaWiFS data during the period 1998-
2007. A comparison of the performances obtained with the two reanalyses demonstrate the benefit of a non-linear analysis scheme that perfoms
anamorphic transformations of the biogeochemical variables. More specifically, the non-linear analysis scheme improves the temporal description
of spatially averaged surface chlorophyll concentration over the whole period of interest decreasing the RMS log error between satellite chloro-
phyll data and model forecasts from 1.25 to 0.74. A comparison of the reanalysed fields colocated in space and time with independent in situ ni-
trate measurements shows that the multivariate scheme allows improving objectively the nitrate concentration estimates in the upper 50 meter sub
-surface layer. In that case the RMS log error is decreased from 0.74 to 0.31 in the first superficial layer of the model. Finally, a number of weak-
nesses are identified in the present products, providing guidance for the tuning of the next reanalysis system.
Introduction
The observation of ocean colour from satellites is a promising source of information to better understand and monitor the ocean biogeochemistry
at global scale. During the past decade, several sensors (e.g. SeaWiFS, MODIS, MERIS) have been deployed to measure globally and daily the
ocean water leaving irradiance at different wave lengths in the visible spectrum. Empirical optical algorithms calibrated according to the sensor’s
properties enable the estimation of biogeochemical quantities such as chlorophyll or chromophoric dissolved organic matter (CDOM) concentra-
tions.
In the framework of the Global Ocean Data Assimilation Experiment (GODAE) program and the EU MyOcean project, a number of research initia-
tives are being conducted to integrate biogeochemical modelling components into state-of-the-art operational global ocean models (Brasseur et
al., 2009). However, since the processes governing the biogeochemistry of the ocean are still poorly known, the coupled physical-biogeochemical
models (CPBM) in place today still exhibit large errors when compared to available field data. Therefore, the optimal combination of ocean colour
observations with biogeochemical models through data assimilation is required to realistically reconstruct the variability of marine productivity in
space and time. Several difficulties must be faced to tackle this challenge, mostly related to the inaccuracy of the data set, the model non-
linearities and the computational resources needed to run coupled experiments. The aim of this study is, on the one hand, to develop the skeleton
of a pre-operational CPBM system with real-time assimilative/forecasting capability, and on the other hand, to operate this prototype system for
producing a biogeochemical reanalysis covering the 1998-2007 period. Such a reanalysis could be used to initialize large-scale CPBMs as well as
to provide realistic boundary conditions for higher-resolution environmental modelling studies at regional scales. Further, the system could also be
used for Observing System Simulation Experiments dedicated to the optimization of observation/sampling strategies (see e.g. Schiller et al.,
2004). So far, the majority of scientific publications dealing with large-scale ocean colour data assimilation have been focused on the description
of idealized twin experiments (e.g., Carmillet et al., 2001; Doron et al., 2011), short-term multivariate experiments (Natvik and Evensen, 2003;
Ourmieres et al., 2009) or long-term univariate experiments (Nerger and Gregg, 2007, 2008). We propose here to perform a realistic, multivariate
data assimilation experiment covering several years, in order to investigate the strength and limitations of a sequential, reduced-order statistical
updating scheme to control the biogeochemical components of a ¼°CPBM of the North Atlantic. To our kn owledge the eddy-admitting resolution
of the model used here makes our investigation unique and fully consistent with the physical models of the Mercator operational suite. This paper
describes first the technical set up of the assimilation experiment variants tested here, focusing on the model features and the key ingredients of
the assimilation system. The impact of the assimilation process on the biogeochemical state variables modeling is then discussed along the simu-
lation period. Finally, we conclude with a number of perspectives to be explored after this first investigation.
A mulƟvariate reanalysis of the North AtlanƟc ocean biogeochemistry during 1998-2007 based on the assimilaƟon of SeaWiFS data
11. #43—October 2011—11Mercator Ocean Quarterly Newsletter
Numerical tools
In this section are described the physical and biogeochemical components of the CPBM configuration, the linear and non-linear updating schemes
implemented in the assimilation setup and the main features and associated errors of the assimilated data.
The physical model
The North Atlantic ocean circulation configuration is based on the NEMO code (Barnier et al., 2006), which is a primitive equation model that uses
the free surface formulation. The prognostic variables are the three-dimensional velocity fields, the thermohaline variables and the sea surface
elevation. Vertical mixing of momentum and tracers is modelled according to the TKE turbulence closure scheme, while convection is parameter-
ized with enhanced diffusivity and viscosity. The model domain covers the North Atlantic basin from 20°S to 80°N and from 98°W to 23°E (Figure
1). Buffer zones are defined at the southern, northern and eastern (Mediterranean) boundaries. The horizontal resolution is ¼°, which is consid-
ered as eddy-permitting in the mid-latitudes. The vertical discretization is done on 45 geopotential levels, with a grid spacing increasing from 6 m
at the surface to 250 m at the bottom. The model is forced by ERA-INTERIM atmospheric data from the ECMWF reanalysis, using bulk formula-
tions. The atmospheric 10-meter humidity, temperature, and wind fields are updated every 6 hours. Short- and long-wave radiations are updated
every days while precipitations are updated every months.
The biogeochemical model
The biogeochemical component of the CPBM is LOBSTER (LOCEAN Biogeochemical Simulation Tool for Ecosystem and Resources), a nitrogen-
based model that contains six prognostic variables: nitrate, ammonium, phytoplankton, zooplankton, detritus and semi-labile dissolved organic
matter. The chlorophyll concentration is considered as a diagnostic variable, estimated using the phytoplankton concentration through a space
and time-dependant Chl/N ratio (Levy et al., 2005). The LOBSTER model updates the biogeochemical concentrations at the same frequency as
the baroclinic physical timestep (i.e. 40 minutes). Biogeochemical variables are then advected and diffused following the oceanic circulation com-
puted by the physical model. A more detailed description of the CPBM can be found in Ourmieres et al. (2009).
The assimilation approach for Ocean Colour Data
As every ocean simulation, the numerical system used for this study is subject to a large variety of modelling errors of different nature. The main
error sources which are suspected to affect both the physical and biogeochemical model response including their mutual coupling, are associated
to imperfect initial conditions and atmospheric fields (themselves derived from a numerical model), parameterizations of unresolved physical or
biological processes (which, in general, involve uncertain parameters or process simplifications), numerical discretization errors, etc. In the pre-
sent implementation however, it is assumed that the ocean colour observations available to improve model estimates only contain the footprint of
errors on the biogeochemistry. As a result, the physical model state is left unchanged by the assimilation of ocean colour data, and the ocean
circulation fields are therefore treated as being assumed perfect.
The assimilation method is implemented using a sequential algorithm derived from the Singular Evolutive Extended Kalman (SEEK) filter (Pham et
al., 1998): the biogeochemical model trajectory is corrected intermittently by computing statistical updates of the state vector using a combination
of available observations and model predictions weighted according to their respective uncertainties. The SESAM tool (Brankart et al., 2011) is
used to perform the algebraic operations related to the assimilation scheme, such as the computation of Empirical Orthogonal Functions (EOFs),
innovation vector or analysis step. The choice of the fixed variant of the SEEK filter (Brasseur and Verron, 2006) is guided by its relative low cost
in terms of numerical resources, an advantage that allows the performance of much longer experiments compared to stochastic ensemble-based
Figure 1: Model domain and biogeochemical regions a, b and c
adapted from the Longhurst (1995) classification.
A mulƟvariate reanalysis of the North AtlanƟc ocean biogeochemistry during 1998-2007 based on the assimilaƟon of SeaWiFS data
12. #43—October 2011—12Mercator Ocean Quarterly Newsletter
methods that require the explicit propagation of estimation errors through the model dynamics (e.g. EnKF). Nevertheless, as ensemble methods
represent promising perspectives in the set-up of operational systems, our assimilation strategy has been carefully chosed to easily allow up-
grades toward ensemble based methods.
The assimilated data set
The data assimilated in this study are chlorophyll concentrations derived from the SeaWiFS sensor using the OC4 algorithm, applied to remotely-
sensed water leaving irradiance measurements (Feldman and McCain, 2010). In order to facilitate the assimilation process, satellite maps were
systematically remapped onto the model grid prior to assimilation. We choose to assimilate the SeaWiFS data set because this sensor is the one
offering the longer time-series of ocean colour data so far (9/1997-12/2010) while others equivalent spatial sensors (i.e. MODIS and MERIS) were
launched later (2002).
The part of water-leaving irradiance measured on top-of-atmosphere by satellite sensors such as SeaWiFS is typically only 5-20% of the total
signal (Lavender et al., 2005). The rest of the signal is due to interactions between light and earth atmosphere (e.g. Rayleigh reflectance, aerosol
reflectance). Therefore in order to obtain accurate satellite data in the visible spectrum, one needs to apply atmospheric corrections to estimate
the water-leaving irradiance. These corrections are a main source of errors when retrieving physical and biological quantities from the original
water leaving irradiance. More specifically, the OC4 algorithm used here contains empirical parameterizations which can yield erroneous estima-
tions of the surface ocean chlorophyll concentration, e.g. due to the assemblage of different phytoplankton types with different optical properties.
Considering this, it is admitted that the root-mean-square log error of SeaWiFS chlorophyll estimation is, on average, 30% of the measured signal
in deep sea waters (Gregg and Casey, 2004).
Protocol of the assimilation experiments
Initialization of the coupled model, simulation period and assimilation setup
The coupled model is initialized on January 1st
, 1998 after a prior spin-up of the circulation model lasting 16 years. The biogeochemical model is
then spun up on a 2 years period starting from the Levitus climatology for nitrate (Garcia et al., 2006) and spatially homogenous values for phyto-
plankton, zooplankton, dissolved organic nitrogen, detritus and ammonium (see Ourmieres et al., 2009). The coupled model is then run without
data assimilation of ocean colour data during a 9 years period until January 1st
, 2007. This simulation will be referred hereafter as the “free” run.
In addition to the free run, two simulations using the data assimilation system are performed. Each of these runs assimilates satellite chlorophyll
maps that are temporally binned every 8 days preceding the analysis date. The state vector used in the assimilation scheme is composed of every
prognostic biogeochemical variable of the three-dimensional grid (phytoplankton, zooplankton, nitrate, ammonium, dissolved organic nitrogen,
detritus). Thus we perform here a multivariate analysis where not only the observed variable (phytoplankton) is corrected but every components of
the biogeochemical model are impacted. Physical variables of the model are not included in the assimilation process and thus are not impacted.
For the first run, the analysis update is performed using the conventional SEEK filter analysis scheme applied to the whole biogeochemical state
vector. This simulation will be referred hereafter as the “linear” run. For the second run, an anamorphosis transformation is applied to the biogeo-
chemical state vector, the observation vector and the error covariance matrix. The general idea is to transform the marginal probability density
functions (pdfs) of the state variables into pdfs that are close to Gaussian. This is achieved by performing a change of variables (anamorphosis)
separately for each single variable of the state vector. The anamorphosis transformation we use here is fully detailed in Béal et al., (2010), while
the effect of this transformation on the representation of model uncertainties is further discussed in Brankart et al. (2011). This simulation will be
referred hereafter as the “anamorphosis” run.
The background error covariance matrix of the analysis scheme is represented using an EOFs basis obtained from the free run variability. For
each analysis date, a new set of EOFs is considered from an ensemble of state vectors spread in time around the analysis date with a 2 days
interval. This ensemble is constructed as follow: for a given day of the year, the model state vectors on a 2 months period surrounding the assimi-
lation date are retained, and that for years 1999 to 2005 (in order not to exceed the free simulation period boundaries). Each of these temporal
ensembles contains 210 members that are used to compute the error covariance matrix by keeping the 20 dominant EOFs. Those EOFs are the
leading directions retained to perform the state vector update. It is noticeable here that the anamorphosis transformation is applied using a tem-
poral ensemble rather than a stochastic ensemble (see below), which has never been tested yet, to our knowledge.
SeaWiFS chlorophyll concentrations are converted into phytoplankton concentration using the Chl/N ratio computed by the coupled model. These
phytoplankton maps are then assimilated in the coupled model and considered as direct measurements of the phytoplankton concentration in the
upper surface layer. The error associated to every single piece of data is set to 30% of the measured signal, in agreement with the admitted Sea-
WiFS averaged error in deep sea waters. The spatial coverage of the data is a common issue for assimilation experiments, raising the question of
A mulƟvariate reanalysis of the North AtlanƟc ocean biogeochemistry during 1998-2007 based on the assimilaƟon of SeaWiFS data
13. #43—October 2011—13Mercator Ocean Quarterly Newsletter
the spatial correlations of the signal. As an example, many extreme latitude regions are not visible by satellite ocean colour sensors during several
month of the year (winter in the northern hemisphere). In those conditions, a local analysis scheme is required to avoid the effect of spurious cor-
relations between mid-latitude and high latitude regions (even though possible connections between distant regions may exist in reality). Consid-
ering this, the choice was made to implement a local analysis scheme with a short influence radius for every distinct data. The horizontal e-folding
radius of influence has been set to 2 grid points and the cut off radius to 5 grid points. This value was chosen as it is compliant with the order of
magnitude of meso-scale features in the mid-latitude regions.
Results and system assessment
Chlorophyll spatial distributions
Figure 2 shows a sequence of surface chlorophyll maps temporally binned on a 60 days period during the year 2006 (first row: day 1 to 60; se-
cond row: day 61 to 120; etc.). These maps are shown for: (a) the free run, (b) the SeaWiFS data, (c) the linear run and (d) the anamorphosis run.
The free run shows a number of systematic differences compared to the SeaWiFS data. The chlorophyll bloom starts slightly too late in the free
run (second row), while the signature of an horizontal structure appears in the Gulf Stream area that is not present in the data (third row). Consist-
ently with this observation, the available nutrients are quickly consumed (not shown) resulting in a strong increase of phytoplankton concentra-
tions. During the peak of the free run bloom, the chlorophyll concentrations seem to be overestimated at high latitudes. When the nutrients are
fully consumed, the chlorophyll concentration decreases quickly (fifth and sixth rows) while the SeaWiFS data exhibit more significant values. To
summarize, the bloom represented by the free run starts too late when compared to satellite data, while nutrients are consumed too quickly. It is
important to notice here that in spite of these differences, the annual chlorophyll evolution is relatively well described (spring bloom, subtropical
gyre). This is an important element to be noticed, as the free run variability provides the basic input to the EOFs computation required by the as-
similation scheme.
Figure 2: Surface chlorophyll distributions
temporally binned on 60 days period for
the year 2006 for: (a) the free run, (b) the
SeaWiFS data, (c) the linear run and (d)
the anamorphosis run. Concentrations
are given in mg(Chl).m-3
.
A mulƟvariate reanalysis of the North AtlanƟc ocean biogeochemistry during 1998-2007 based on the assimilaƟon of SeaWiFS data
14. #43—October 2011—14Mercator Ocean Quarterly Newsletter
When looking at the assimilation results, the bloom starts at a more realistic date, and the horizontal pattern discussed here above (Gulf Stream)
is not present anymore (c and d). During the bloom, the modelled values are in agreement with the data, both in the linear and the anamorphosis
experiments (third and fourth row). However, the chlorophyll values are still underestimated by the end of the year, like in the free run.
The error covariance matrix is computed with free run model outputs on a 2 months period running windows over years. Thus if the free run varia-
bility is much lower than the data variability on a 2 months period, the EOFs set does not contain the intrinsic dynamics allowing to satisfyingly
correct the instantaneous state. This is actually what happen by the end of the year (fifth and sixth rows) when comparing the free run (a) to satel-
lite data (b). This issue could be overcome by increasing the temporal windows on which the EOFs are computed. This will allow to catch more
variability in the EOFs set used for the SEEK analysis. Nevertheless ones must be aware that increasing this temporal window could also lead to
erroneous corrections as it will include in the correction base past or future correlations, not present in the instantaneous system state. From a
stochastic point of view, it will include in the correction base numerous states having an extremely low probability to happen (on the edges of the
temporal window).
The impact of the assimilation scheme is weaker in coastal waters; this is a somehow expected result as the NEMO model is not fine-tuned to
describe precisely small scale oceanic circulation features in the coastal zone. In addition, river plumes are simply considered as runoffs with no
nutrient input associated to the river discharges. These processes, however, can play a major biogeochemical role in the coastal area, as demon-
strated in the context of ocean colour data assimilation experiments by Fontana et al. (2009, 2010).
Some spots of high chlorophyll concentration appear in the linear run (c; 0°-15°N; fourth row) while these patterns are not present either in the
other run, nor in the data. This is a likely consequence of the multivariate analysis that is performed here. In highly productive zones, such as in
front of the Senegal Rivers (15°N, 25°W) where upwe lling and nitrate inputs govern the biogeochemical response, it may happen that erroneous
corrections applied to non-observed variables (e.g. nutrients) result into spurious phytoplankton growths. These patterns do not appear in the
anamorphosis run (d), attesting that the corrections applied to non-observed variable are more realistic when using a non-linear assimilation
scheme.
Chlorophyll temporal evolution
Figure 3 shows the temporal evolution of the horizontally averaged chlorophyll concentration in the first layer of the coupled model over the simu-
lation period. Both data and model outputs were temporally binned into 16-day periods. The temporal evolution is shown for 3 open-sea biogeo-
chemical regions (a, b and c; see Figure 1) adapted from the Longhurst (1995) classification. The green fringe under the SeaWiFS curve is an
indicator of the satellite data spatial coverage over the considered region. The fringe thickness is zero when the whole considered region is cov-
ered by the satellite data. Then the fringe thickness increases linearly with respect to the number of missing observations to reach 100% of the
considered data value when almost no data are available. Of course when no data are present at all, no dots and no fringe are drawn.
Figure 3: Temporal evolution of the horizontally
averaged chlorophyll concentration in the first
layer of the coupled model for the different bio-
geochemical regions (see Figure 1). The chloro-
phyll concentration is expressed in mg(Chl).m-3
.
A mulƟvariate reanalysis of the North AtlanƟc ocean biogeochemistry during 1998-2007 based on the assimilaƟon of SeaWiFS data
15. #43—October 2011—15Mercator Ocean Quarterly Newsletter
It appears on Figure 3 that for high-latitude regions (a) the bloom as modelled by the free run starts too late, does not last as long as it should
when compared to data, and values are overestimated at the peak of the bloom, as discussed in the previous paragraph. One can see that, during
the bloom period, the spatial SeaWiFS coverage is high (small green fringe) allowing a well-controlled chlorophyll concentration for both linear and
anamorphosis assimilation runs. It is unfortunately not the case in winter when almost no data are present during several months of the year for
high latitude regions. Thus in winter period and for high latitudes, it appears that not enough data are integrated in the local assimilation process to
allow a significant improvement of the mean surface chlorophyll. For the mid-latitude region b, the bloom period is well reproduced by the free run,
while the data are largely overestimated during the peak of the bloom, mainly due to the spurious Gulf Stream pattern discussed before. It is
shown that, for this region, the linear and anamorphosis runs are able to efficiently constrain the chlorophyll evolution. Further, a drift of the chloro-
phyll concentration appears after several years of free simulation while the signal is more stable when using the assimilation process. For the mid-
latitude region c, the free run performs well both in terms of timing and maximum value of the bloom. The anamorphosis run outstandingly im-
proves the chlorophyll description all along the simulation period. The results are totally different when considering the linear run. Indeed, after the
first year of simulation, the chlorophyll significantly diverges from the data and the others simulations. This could be the consequence of the high
chlorophyll spots appearing in that region as described in the previous paragraph. The unrealistic corrections applied on non-observed variables in
the case of the linear run are responsible for side effects on the biogeochemical dynamics that make the method fail for pluri-annual simulations.
This demonstrates how the anamorphosis transformation can be a critical point in setting up an efficient and stable assimilation system for CPBMs
using satellite ocean colour data.
Surface chlorophyll concentrations forecast
During the assimilation period, every distinct assimilated observation was compared to its 8-days forecast equivalent for each modeled situation.
This comparison is based on 410 SeaWiFS images and more than 1.5 107
distinct data. Figure 4 shows the probability density function (pdf) of the
log(CSeaWiFS/Cmodel) function where CSeaWiFS and Cmodel stand for the SeaWiFS and the model concentration (resp.). To compute this pdf, only open
sea data entered the computation. Open sea regions were defined as area with a bathymetry deeper than 500 meters.
It appears on Fig. 4 that the chlorophyll content forecasting is strongly biased by the free run, with an overestimation of the measured values by
the model (negative values) implying a RMS of 1.25. This general behavior is mainly due to the overestimation of the maximum chlorophyll con-
centrations in high latitudes regions during the spring bloom, as discussed before. The linear run shows a significant improvement of this compari-
son by centering the pdf closer to 0 while a light trend to overestimation is still visible. In the case of the linear run the computed RMS was 0.86.
The result is even better when considering the non-linear run. Indeed, while the non-linear pdf is unchanged for extreme values (log error lower
than -2 and higher than 2) when compared to the linear one, the pdf is better centered on 0 for low values of log error. Concerning the non-linear
RMS, result was slightly better than the linear run with a value of 0.74 and thus a reduction of 41% compared to the free run RMS.
Comparison of nitrate concentration with an independent data set
The World Ocean Atlas 2009 (WOA09) data set has been used to assess and validate the assimilation experiments with respect to independent,
unassimilated in situ measurements of nitrate. The historical data reported in this atlas were obtained from the National Oceanographic Data Cen-
ter (NODC) and the World Data Center (WDC) archives that include the measurements gathered as part of the GODAR and WOD project (Boyer
et al., 2006). The horizontal, vertical and temporal coverage of the available data set allows the comparison with the outputs of the 3 simulations
(free, linear, anamorphosis) colocated in space and time.
Figure 4: Probability density func-
tion of the log(CSeaWiFS/Cmodel)
function computed for every dis-
tinct forecasted satellite data.
A mulƟvariate reanalysis of the North AtlanƟc ocean biogeochemistry during 1998-2007 based on the assimilaƟon of SeaWiFS data
16. #43—October 2011—16Mercator Ocean Quarterly Newsletter
The data were grouped on running vertical intervals (0-5; 5-10; 10-50; 30-70; 50-90; 70-110; 90-130; 110-150; 130-170, in meters). On these data
groups, the logarithmic Root Mean Square error was computed as follows:
where Cmod and Cin situ stand for the modelled concentration and the measured concentration (resp.) and N is the number of data on which the
computation is done. For each depth interval listed above N is equal to (resp.): 357, 286, 1100, 886, 745, 1544, 1554, 487, 62.
Figure 5 shows the vertically distributed RMS for all available nitrate data higher than 1 mmol.m-3
contained in the WOA09. The RMS curves are
shown for the three situations: free, linear, and anamorphosis. It appears that the assimilation of satellite chlorophyll data reduces the error in the
upper layers of the water column, up to the 10-50m interval. While both the linear and anamorphosis runs improve the accuracy of the nitrate esti-
mates in the upper part of the water column, the anamorphosis performs much better than the linear one. In the anamorphosis run, the RMS is
decreased from 0.74 to 0.31 mmol.m-3
in the 0-5m interval. The scatter plot shown in a frame represents the model versus in situ values in upper
10m layer (same colours as the RMS plot). It is shown that a number of nitrate values, largely underestimated in the free run (black dots in the
frame of Fig. 5), are satisfyingly corrected in the linear run. On the other hand, large overestimations appear in the linear run case, especially for
values close to 1 mmol.m-3
(blue dots in the frame of Fig. 5). This overestimation of nitrate is a possible explanation for secondary effects dis-
cussed above on chlorophyll dynamics at mid-latitudes regions. The anamorphosis run corrects underestimations in the same way that the linear
run but overestimations of values close to 1 mmol.m-3
disappear (red dots in the frame of Fig. 5), explaining the good performance of the method
on the 0-5 and 5-10 m depth interval with respect to the RMS. On the deeper part of the euphotic layer (from about 50 to 150 m.), the linear run
clearly increases the RMS when compared to others situations. Efficiency of the anamorphosis run is mixed although differences with the free run
remain nevertheless weak. The RMS profiles shown on Figure 5 attest that, in the context of this study, the assimilation of sea surface chlorophyll
concentration is not sufficient to control the open ocean biogeochemistry in the water column below some tens of meters depth.
As already stated, the comparison discussed above was made with available nitrate data for measured concentrations higher than 1 mmol.m-3
.
Concerning the data lower than 1 mmol.m-3
, the results did not show clear improvements due to the assimilation. There are several reasons that
could explain that the assimilation process does not perform well for this range of data. First, these data are generally measured in the oligotrophic
∑=
−=
N
i
situin
N 1
2
10mod10 ))C(log)C((log
1
RMS
Figure 5: Nitrate RMSlog misfit in the free,
linear and anamorphosis runs, as a function
of depth (in meters) estimated using all avail-
able in situ measurements higher than
1 mmol.m-3
contained in the WOA09 data
set. Frame: model versus in situ scatter plot
for the depth interval 0-10m. Concentrations
are given in mmol(NO3).m-3
.
A mulƟvariate reanalysis of the North AtlanƟc ocean biogeochemistry during 1998-2007 based on the assimilaƟon of SeaWiFS data
17. #43—October 2011—17Mercator Ocean Quarterly Newsletter
sub-tropical gyre, where the low temporal variability of biogeochemical concentration for the free run (used to compute the EOFs) does not allow
the assimilation system to correct strong differences between model and data. Secondly, these data are also generally measured in strong nutri-
ent gradients, i.e. in a transition zone between nutrients enriched waters (high latitudes) and oligotrophic waters of the subtropical gyre. In this
region a well-reproduced physics is essential in order to obtain a satisfying biogeochemical dynamics. As the physics was not constrained by the
assimilation process, this transition zone may not be located exactly preventing the assimilation system to perform correctly.
Conclusion
Realistic ocean colour assimilation experiments have been performed in a CPBM of the North Atlantic basin as part of the GMMC Mercator-Vert
and FP7 MyOcean projects. These experiments aimed at assessing the feasability of multi-year reanalyses of the ocean biogeochemistry based
on SeaWiFS ocean colour data in a ¼° eddy-admitting model. Different simulations were conducted during a 9 year-long period (1998-2007),
allowing the assimilation of 410 SeaWiFS-estimated maps of chlorophyll temporally binned every 8 days. The assimilation was performed with a
fixed variant of the reduced-rank Kalman filter (SEEK) in order to limit the computational burden of the data integration process. Several key pa-
rameters entering the analysis step (e.g. model and observation errors parameterizations, local influence radius) were carefully tuned to maximize
the benefit of the assimilation process. Comparisons were made between a free run, an assimilation run using a linear updating scheme, and an
assimilation run using a non-linear updating scheme with anamorphic transformations.
Overall, the assimilation experiments demonstrate that the SeaWiFS data provide a very powerfull means of controlling the seasonal evolution of
primary production in the North Atlantic ocean. The basin-averaged surface chlorophyll concentration was computed over several regional do-
mains inside the modelled area. The results showed that the use of a non-linear analysis scheme largely improved the description of this variable
along the simulation period. The link between data spatial coverage and assimilation efficiency was underlined for high latitude regions. Neverthe-
less, unrealistic corrections were detected locally in case where a linear scheme was used to assimilate the data. These anomalies appeared to
be the consequence of spurious multivariate corrections of non-observed variables.
The assimilation impact on nitrate variables was assessed using the WOA09 data set. Spatial and temporal colocations were performed between
the model outputs and a total of 4110 in situ data during the simulation period. The RMS misfit between the assimilation products and unassimilat-
ed in situ measurements was computed within several vertical intervals of the water column. It appears that the description of the nitrate concen-
tration was improved between the surface and 50 m depth when using the non-linear assimilation scheme. Compared to the free run, the linear
scheme improved the surface nitrate values too but a trend to overestimation was observed between models and data in that case. Differences
between the free run and the non-linear scheme in the 50-150 m depth interval were rather small, indicating that the assimilation of surface chloro-
phyll data might not be sufficient to control the CBPM dynamics over the whole euphotic layer. Concerning the deeper part of the water column,
the use of a linear scheme clearly increased the nitrate RMS, while the non-linear scheme did not show such a negative impact.
This study opens new perspectives for large-scale biogeochemical simulations assimilating ocean colour data. Firstly, the pre-operational system
used here demonstrates that the method is efficient and stable, paving the way for robust real-time biogeochemical nowcast/forecast in the near
future. Secondly, the computation of more realistic biogeochemical climatologies and open-sea boundary conditions can be expected from this
kind of multi-annual reanalyses. This is a much expected opportunity since the currently available nutrient climatologies based on objective data
interpolation are still hampered in a number of ways by the lack of in situ data (Garcia et al., 2006).
A number of improvements of the system presented in this article will have to be explored before undertaking operational transition into the Merca-
tor-Ocean suite. As already stated before, the physical simulation was not constrained by any data assimilation scheme in the experiment de-
scribed in this paper. Constraining the physical variables should improve both the biogeochemical dynamics but also the ocean colour data assim-
ilation efficiency through a better description of the error sources. Currently, an operational simulation is run by the Mercator-Ocean group where
the physics is constrained by an assimilation process while the biogeochemical variables evolve freely (so-called BIOMER1V1). Merging the BIO-
MER1V1 system with the ocean colour data assimilation system described here represents the future of the global ocean biogeochemistry fore-
casting.
Nevertheless some extensive studies should be performed prior to reach this milestone with regard to the biogeochemical assimilation process.
Sensitivity experiments should be intensified to refine the parameterization of model errors (e.g. through ensemble-based approaches) and obser-
vation errors, as well as for improving the localization setup. An obvious limitation considered in the present approach arises from the assumption
that satellite-derived chlorophyll concentrations are confined to the surface level of the CBPM, while the satellite measured values are actually
integrating a finite portion of the euphotic depth. Therefore, considering the verticality of ocean colour measurements could be a more sensible
way for combining ocean colour data with biogeochemical models.
Acknowledgement
This work was conducted as a contribution to the MyOcean Project funded by the E.U. (Grant FP7-SPACE-2007-1-CT-218812-MYOCEAN), with
A mulƟvariate reanalysis of the North AtlanƟc ocean biogeochemistry during 1998-2007 based on the assimilaƟon of SeaWiFS data
18. #43—October 2011—18Mercator Ocean Quarterly Newsletter
additional support from the Groupe Mission Mercator Coriolis (GMMC) as part of the Mercator-Vert project. The calculations were performed using
HPC resources from GENCI-IDRIS (Grant 011279).
References
Barnier, B., Madec, G., Penduff, T., Molines, J.M., Treguier, A.M., Le Sommer, J., Beckmann, A., Biastoch, A., Boning, C., Dengg, J., Derval, C.,
Durand, E., Gulev, S., Remy, E., Talandier, C., Theetten, S., Maltrud, M., McClean, J., and De Cuevas, B., 2006. Impact of partial steps and mo-
mentum advection schemes in a global ocean circulation model at eddy-permitting resolution. Ocean Dynamics, 56: 543–567. doi: 10.1007/
s10236-006-0082-1.
Boyer, T., Antonov, J., Garcia, H., Johnson, D., Locarnini, R., Mishonov, A., Pitcher, M., Baranova, O., and Smolyar, I., 2006. World Ocean Data-
base 2005. U.S. Government Printing Office, Washington D.C., 190 pp.
Brankart J.-M., Testut C.-E., Béal D., Doron M., Fontana C. Meinvielle M. and Brasseur P., 2011 : Towards an improved description of oceano-
graphic uncertianties : effect of local anamorphic transformations on spatial correlations, submitted to Ocean Sci.
Brasseur P. and Verron J., 2006 : The SEEK filter method for data assimilation in oceanography: a synthesis, Ocean Dynamics, 56, 650-661,
doi :10.1007/s10236-006-0080-3.
Brasseur, P., Gruber, N., Barciela, R., Brander, K., Doron, M., El Moussaoui, A, Hobday, A.J., Huret, M., Kremeur, A.S., Lehodey, P., Matear, R.,
Moulin, C., Murtugudde, R., Senina, I., and Svendsen, E., 2009. Integrating Biogeochemistry and Ecology Into Ocean Data Assimilation Systems.
Oceanography, 22(3): 206–215.
Béal, D., Brasseur, P., Brankart, J.M., Ourmières, Y., and Verron, J., 2010. Characterization of mixing errors in a coupled physical biogeochemical
model of the North Atlantic : implications for nonlinear estimation using Gaussian anamorphosis. Ocean Science, 6(1): 247–262. doi :10.5194/os-6
-247-2010.
Carmillet, V., Brankart, J.M., Brasseur, P., Drange, H., Evensen, G., and Verron, J., 2001. A singular evolutive extended Kalman filter to assimilate
ocean colour data in a coupled physical-biochemical model of the North Atlantic ocean. Ocean Modelling, 3(3-4) : 167 – 192. doi: 10.1016/S1463-
5003(01)00007-5.
Doron, M., Brasseur, P., and Brankart, J.M., 2011. Stochastic estimation of biogeochemical parameters of a 3D ocean coupled physical-
biogeochemical model : Twin experiments. J. Mar. Sys., 87(3-4): 194 – 207. doi: 10.1016/j.jmarsys.2011.04.001.
Feldman, G., McCain, C., 2010. Ocean colour web, seawifs reprocessing 2009.1. NASA Goddard Space Flight Center. Eds. Kuring, N., Bailey, S.
W., http://oceancolor.gsfc.nasa.gov/.
Fontana, C., Grenz, C., Pinazo, C., Marsaleix, P., and Diaz, F., 2009. Assimilation of SeaWiFS chlorophyll data into a 3D-coupled physical bioge-
ochemical model applied to a freshwater-influenced coastal zone. Continental Shelf Research, 29(11-12) : 1397 – 1409. doi: 10.1016/
j.csr.2009.03.005.
Fontana, C., Grenz, C., and Pinazo, C., 2010. Sequential assimilation of a year-long time-series of SeaWiFS chlorophyll data into a 3D biogeo-
chemical model on the French Mediterranean coast. Continental Shelf Research, 30(16): 1761 – 1771. doi: 10.1016/j.csr.2010.08.003.
Garcia, H., and. T.P. Boyer, R.L., and Antonov, J., 2006. World Ocean Atlas 2005, Volume 4 : Nutrients (phosphate, nitrate, and silicate). U.S.
Government Printing Office, Washington D.C., 396 pp.
Gregg, W.W. and Casey, N.W., 2004. Global and regional evaluation of the SeaWiFS chlorophyll data set. Remote Sensing of Environment, 93
(4): 463 – 479. doi: 10.1016/j.rse.2003.12.012.
Lavender, S., Pinkerton, M., Moore, G., Aiken, J., and Blondeau-Patissier, D., 2005. Modification to the atmospheric correction of SeaWiFS ocean
colour images over turbid waters. Continental Shelf Research, 25(4): 539 – 555. doi: 10.1016/j.csr.2004.10.007.
Longhurst, A., 1995. Seasonal cycles of pelagic production and consumption. Progress In Oceanography, 36(2): 77 – 167. doi: 10.1016/0079-
6611(95)00015-1.
Lévy, M., Gavart, M., Mémery, L., Caniaux, G., and Paci, A., 2005. A four-dimensional mesoscale map of the spring bloom in the northeast Atlan-
tic (POMME experiment): Results of a prognostic model. J. Geophys. Res., 110(C7): C07S21.
A mulƟvariate reanalysis of the North AtlanƟc ocean biogeochemistry during 1998-2007 based on the assimilaƟon of SeaWiFS data
19. #43—October 2011—19Mercator Ocean Quarterly Newsletter
Natvik, L.J. and Evensen, G., 2003. Assimilation of ocean colour data into a biochemical model of the North Atlantic : Part 1. Data assimilation
experiments. Journal of Marine Systems, 40-41: 127 – 153. doi: 10.1016/S0924-7963(03)00016-2. The Use of Data Assimilation in Coupled Hy-
drodynamic, Ecological and Bio-geo-chemical Models of the Ocean. Selected papers from the 33rd International Liege Colloquium on Ocean Dy-
namics, held in Liege, Belgium on May 7-11th, 2001.
Nerger, L. and Gregg, W.W., 2007. Assimilation of SeaWiFS data into a global ocean-biogeochemical model using a local SEIK filter. Journal of
Marine Systems, 68(1-2): 237 – 254. doi: 10.1016/j.jmarsys.2006.11.009.
Nerger, L. and Gregg, W.W., 2008. Improving assimilation of SeaWiFS data by the application of bias correction with a local SEIK filter. Journal of
Marine Systems, 73(1-2): 87 – 102. doi: 10.1016/j.jmarsys.2007.09.007.
Ourmières, Y., Brasseur, P., Lévy, M., Brankart, J.M., and Verron, J., 2009. On the key role of nutrient data to constrain a coupled physical-
biogeochemical assimilative model of the North Atlantic Ocean. Journal of Marine Systems, 75(1-2): 100 – 115. doi:10.1016/
j.jmarsys.2008.08.003.
Pham, D.T., Verron, J., and Roubaud, M.C., 1998. A singular evolutive extended Kalman filter for data assimilation in oceanography. Journal of
Marine Systems, 16(3-4): 323 – 340. doi: 10.1016/S0924-7963(97)00109-7.
Schiller, A., Wijffels, S.E., and Meyers, G.A., 2004. Design Requirements for an Argo Float Array in the Indian Ocean Inferred from Observing
System Simulation Experiments. Journal of Atmospheric and Oceanic Technology, 21(10): 1598–1620. doi: 10.1175/1520-0426(2004)021.
A mulƟvariate reanalysis of the North AtlanƟc ocean biogeochemistry during 1998-2007 based on the assimilaƟon of SeaWiFS data
20. #43—October 2011—20Mercator Ocean Quarterly Newsletter
IFREMER/CERSAT ARCTIC SEA ICE DRIFT 1992-2010 TIME SERIES FROM
SATELLITE MEASUREMENTS FOR MYOCEAN
By Fanny Girard-Arhuin(1)
, Denis Croize-Fillon(2)
(1)
IFREMER, Laboratoire d'Océanographie Spatiale, Plouzané, France
(2)
IFREMER, CERSAT, Plouzané, France
Abstract
Satellites enable daily and global coverage of the polar oceans, providing a unique monitoring capability of sea ice dynamics. Sea ice drift fields
can be estimated in Arctic from several sensors, in particular scatterometers and radiometers. The 1992-2010 winter sea ice drift time series
made at Ifremer/CERSAT and available for the GMES-MyOcean project (http://www.myocean.eu) are presented here. There are several prod-
ucts : the Level 3 products from single sensors and the L4 products from the combination of sensors. They are available at 3, 6 and 30 day-lag
with a 62.5 km-grid size during winter. Results of the validation with in situ data are shown. Drift data density shows that almost full maps are
available over the winter thanks to the merging of the data and the interpolation of the missing points. These data are available for operational use
and the scientific community.
Introduction
Sea ice cover and motion have major impacts on heat fluxes between ocean and atmosphere in polar areas. Moreover, the Arctic ice growth and
melt impact the fresh water flux which then has an important role in the thermohaline circulation, in particular in European seas (Aagaard et al.,
1985). This is why sea ice observations are required, in particular long time series in order to better understand the role of sea ice in ocean circu-
lation or climate change.
The sea ice drifts shown here concern only the Arctic ocean where it has been extensively validated with in situ data. Since 1979, buoys are
moored in Arctic ice, each year, in the framework of the International Arctic Buoy Program (IABP). They provide continuous local measurements
(pressure, air temperature, ice drift) covering mostly the Central Arctic and Beaufort Sea, but the spatial distribution is rather sparse in the Eura-
sian Seas, providing a limited view of the velocity field (Colony and Thorndike, 1984). Since the 1990’s, sea ice drift can be estimated from satel-
lite data, daily and with global coverage of the polar ocean. Given the sensor spatial resolution and the magnitude of the expected drifts, the sig-
nature of the feature to be tracked must persist for several days. The ice motion is thus estimated at the scale of large ice floes, whereas a buoy
measures the drift of a single ice floe. Data from high spatial resolution satellite sensors have been used for sea ice drift estimate in feasibility
studies, for example with Advanced Very High Resolution Radiometer (AVHRR, Emery et al., 1991) and Synthetic Aperture Radar (Kwok et al.,
1990). These demonstrations are limited to local or regional studies. Here, we focus on sea ice drift at large scale (or polar scale). These esti-
mates are inferred from passive (radiometers) and active microwave sensors (scatterometers). High frequencies data are very sensitive to atmos-
pheric effects (moisture and liquid water). This implies sea ice drift estimations at periods of low water vapour during winter, from October to April
(for example Martin and Augstein, 2000).
This paper presents some Arctic sea ice drift datasets inferred from satellite measurements; they are produced daily in winter at Ifremer via the
Centre d’Exploitation et de Recherche SATellitaire (CERSAT), and are available as a time series for the MyOcean project. The datasets present-
ed here are the level 3 products corresponding to the daily Arctic drift data at 3 and 6 day-lags from single sensors, and the level 4 products corre-
sponding to the merging of radiometer and scatterometers sea ice drift fields. These time series cover the period 1992-2010 and are available
from October until April (cold period). Section 1 presents the satellite sensors used, Section 2 the sea ice drift fields (level 3 products), Section 3
the merging of sea ice drift fields (level 4 products). Applications of these datasets are in Section 4, conclusions follow in Section 5.
Sensors
In this section, we present briefly the passive and active microwave sensors onboard satellites we use to infer the sea ice drift fields.
Ifremer/CERSAT ArcƟc sea ice driŌ 1992-2010 Ɵme series from satellite measurements for myocean
21. #43—October 2011—21Mercator Ocean Quarterly Newsletter
Radiometers
Since the 1980’s, passive microwave radiometers onboard satellites as the Special Sensor Microwave Imager(s) (SSM/I) have been commonly
used to estimate sea ice concentration from the daily brightness temperature data. The pixel resolution is 25 km at the lower frequencies and 12.5
km with the 85.5 GHz channel, we focus here on the 12.5 km resolution data which starts in 1992.
Passive microwave radiometers data like SSM/I brightness temperature data are commonly used to estimate sea ice concentration, they also
have been widely used to estimate Arctic sea ice drift. In order to determine sea ice drift from successive brightness temperature maps, stable
radiation of the sea ice cover and negligible influence of the atmospheric conditions are needed.
Scatterometers
The benefit of the all-weather, day-night microwave radar measurements has been well established. During the 1999-2009 period, daily averaged
backscatter maps at the resolution of 12.5 km have been built from the Ku-band SeaWinds/QuikSCAT scatterometer data. Since 2007, ASCAT
scatterometer onboard MetOp provides C-band backscatter data we process to have daily averaged backscatter maps (same pixel size resolu-
tion). These data are used to infer sea ice drift fields.
Single sensors sea ice drift fields
Sea ice drift fields at global scale exist since the beginning of the Arctic monitoring with the low frequency channels of the SSM/I radiometer sen-
sor. In this section, some methods to infer sea ice drifts are briefly presented, and the one applied at Ifremer/CERSAT highlighted. Results of vali-
dation with in situ buoys data follow.
Methods
All methods assume that the structures tracked have spatial dimensions larger than the pixel resolution. For the pixel resolution available, single
ice floes cannot be detected. Several methods have been tested to determine sea ice displacements : algorithms are based on tracking common
features in pairs of sequential satellite maps.
One tool which is the most widely used is the Maximum Cross Correlation (MCC), it was used successfully with SSM/I or AVHRR data for exam-
ple (Emery et al., 1991; Ninnis et al., 1986). This method enables detection of translation displacement and no rotation (Kamachi, 1989; Ninnis et
al., 1986). A correlation is estimated at each position between two arrays: one at a given day and another one lagged in time in order to obtain a
correlation coefficient array. The location of the maximum correlation is the location of the maximum similarity between the two original images.
The displacement can thus be inferred. Details about tracking methods can be found among others in Ezraty et al. (2007a,b), Martin and Augstein
(2000), Emery et al. (1997), Kwok et al. (1998), and in Lavergne et al. (2010) who introduce recently an alternative tracking method based on
MCC.
For the products presented here, we apply the MCC tool to the Laplacian field of brightness temperature or backscatter data (from radiometer and
scatterometer sensor respectively) in order to enhance the structures to be tracked. In order to remove outliers, a minimum coefficient correlation
is imposed. In addition, a comparison with the wind pattern is applied (here ECMWF model wind field, NCEP re-analyses from Kwok et al., 1998)
since mean sea ice drift is strongly linked with the geostrophic wind (Thorndike and Colony, 1982).
Another tool to enhance the feature to be tracked is the wavelet analysis. It is applied on various spatial scales in order to separate ice features. A
two-dimensional Gaussian wavelet (also called Mexican hat wavelet) has been used for ice feature detection by Liu and Cavalieri (1998) and Liu
et al. (1999). It turns out that this wavelet technique is very similar to the Laplacian field enhancement as used for the Ifremer/CERSAT product.
Results
Some results of SSM/I drift validation are summarized in Table 1. The displacements (in km) are usually converted into ice speed (in cm s-1
) in
order to compare the quality of the drift magnitude estimated at different day-lags or at various pixel size. Nevertheless, this introduces a non line-
ar effect in the ratio. Moreover, the constant half-pixel minimum displacement embedded in the original measurement is then erroneously account-
ed for.
These results show that with SSM/I, sea ice drifts can be retrieved with good accuracy. Note that with a pixel resolution of 12.5 km, for 1 day-lag,
the minimum displacement that can be detected is 6.25 km which corresponds to a velocity of 7 cm s-1
, whereas the mean buoys velocity are
about 7 cm s-1
. Then the day-lag value to be used for a 12.5 km resolution pixel must be larger: 3 day-lag corresponds to a minimum velocity of
4.8 cm s-1
. The 3-day lag is commonly used. The day-lag must be chosen accordingly to the magnitude of the expected drift (small drift – large
Ifremer/CERSAT ArcƟc sea ice driŌ 1992-2010 Ɵme series from satellite measurements for myocean
22. #43—October 2011—22Mercator Ocean Quarterly Newsletter
lag; large drift – short lag). That’s why the Ifremer/CERSAT products used 3 day-lag and 6 day-lag to infer large and small drifts (Ezraty et al.,
2007a). The time series of these drift inferred from SSM/I channels are available since 1992. A 2 day-lag is also used with lower resolution pixel
data like the AMSR-E radiometer data (Ezraty et al., 2007c, Lavergne et al., 2010).
The SeaWinds/QuikSCAT and the ASCAT/MetOp scatterometer daily averaged backscatter maps are also used at Ifremer/CERSAT to estimate
sea ice drift fields applying the same method at 3 and 6 day-lag (Ezraty et al., 2007b). The SeaWinds/QuikSCAT drift field time series is available
for the period 1999-2009, the ASCAT/MetOp time series has started in 2007 and is ongoing. Figure 1 shows the drift data density during the 2003-
2004 winter. Radiometer channels (SSM/I H & V) provides 60 to 75% data density during the winter, less at the beginning of the winter, which is
the worst period of estimation. QuikSCAT data density has a slightly higher density, in particular during the January-April period with a density
reaching about 80%.
Sea ice drifts deduced from radiometers have a reasonable accuracy but are limited by data gaps and low data density at the beginning and
sometimes also at the end of the cold period. This can be improved using scatterometer data to built a merged product.
Merging sea ice drift fields
Due to satellite geometry and near-polar orbit, the data gap near the North Pole for the scatterometers is smaller than that of the SSM/I. Moreover,
radars at Ku-band do not have the problem caused by water vapour as the 85.5 GHz SSM/I channels, so QuikSCAT is an adequate sensor to fill
in poorly tracked areas where the weather influence prevents reliable SSM/I estimation during the cold season. We present here the advantages
brought by the combination of several independent drift fields from radiometer and scatterometer as performed at Ifremer/CERSAT, more details
can be found in Girard-Ardhuin et al. (2008).
Sea ice drift maps from SSM/I are based on the two channels of the sensor (horizontal and vertical polarizations, ‘H’ and ‘V’): from each channel,
a drift map is inferred. This results in two independent drift fields. Merging the two fields gives a drift product which is more reliable than the one
inferred from a single channel. We also combine these two fields and the QuikSCAT sea ice drift field by optimal merging, leading to the “Merged”
product.
The Merged product has been validated using IABP buoys during five winters (see Table 2): the standard deviation of the difference at 3 day-lag
is 2.91 cm s-1
, comparable with previous results quoted above for 3 and 4 day-lags (Table 1). Using 6 day-lag is even more adequate for small
std speed differ-
ence
(cm s-1
)
std direcƟon
difference
(°)
day-lag
(nb of days) reference
6 1 Emery et al., 1997
2.6 à 2.9 18 à 25.9 1 Liu and Cavalieri, 1998
4.28 4.52
2.58 * 2.89 *
46.5 50.4
30.6 * 32.8 *
3 Kwok et al., 1998
2.96 34.4 4 Liu et al., 1999
2.27 * 35.5 * 4 Zhao et al., 2002
Table 1—Results of the comparison
between SSM/I drift vectors and
buoys drifts for several experi-
ments : standard deviation (std) for
sea ice drift difference and for direc-
tion difference, number of day-lag,
and reference. * indicates that the
drift magnitudes less than one pixel
are excluded.
std speed differ-
ence
(cm s-1
)
std direcƟon
difference
(°)
day-lag
(nb of days)
2.91
2.90 *
39.2
29.6 *
3
1.72
1.76 *
29.6
24.3 *
6
Table 2— Results of the comparison between Ifremer/CERSAT
“Merged” SSM/I/QuikSCAT drift vectors and buoys drifts for five
winters : standard deviation for sea ice drift difference and for
direction difference, and number of day-lag. * indicates that the
drift magnitudes less than one pixel are excluded.
Ifremer/CERSAT ArcƟc sea ice driŌ 1992-2010 Ɵme series from satellite measurements for myocean
23. #43—October 2011—23Mercator Ocean Quarterly Newsletter
drifts with a standard deviation of the difference of 1.72 cm s-1
. Consequently, the angular data have a better resolution. The angle difference
sharply decreases (to values lower than 45°) for dr ifts higher than 40 km (also noticed by Liu et al., 1999). Standard deviations decrease when
small drifts (displacement lower than the pixel size) are excluded, in particular for drift direction.
The combination of the independent drift fields provides better confidence in the final resulting field than for the individual ones since each drift is
inferred from independent measurement. First, the data gap at the North Pole is reduced to the smallest one. Second, the sea ice Merged drift
field enables discrimination of remaining vectors outliers of the individual products. Third, the number of valid drift vectors is higher than the single
sensors valid drift vectors: the merging increases the data density up to 90% for December-April period and more than 80% at fall in the example
of figure 1.
Figure 1 : Times series of drift
data density (percentage x 100)
for the 2003-2004 winter for
SSM/I at horizontal (in green)
and vertical (in blue) polariza-
tions, QuikSCAT (in black),
Merged SSM/I/QuikSCAT (in
red) and filled Merged SSM/I/
QuikSCAT (in brown) drifts at 3
day-lag.
Figure 2: Merged filled Arctic
sea ice drift field from SSM/I
radiometer drift fields and Sea-
Winds/QuikSCAT scatterometer
drift field at 3 day-lag on October
3rd
-6th
, 2007. Grid spacing is
62.5 km. Drift vectors less than
one pixel are marked with a
cross.
Ifremer/CERSAT ArcƟc sea ice driŌ 1992-2010 Ɵme series from satellite measurements for myocean
24. #43—October 2011—24Mercator Ocean Quarterly Newsletter
In order to ease the use of these datasets by modelers, we have developed a space and time interpolation to infer drift in some data gaps patch-
es. Figure 2 shows an example of this filled drift field at 3 day-lag in October, at the beginning of freeze period, with an almost full drift map, thanks
to the interpolation which enables to increase by 5 to 10% the drift data density (also see in Fig. 1 in brown). This space and time interpolation is
very useful, in particular for fall estimation when drift is more difficult to infer.
Applications
Drift datasets are useful for several applications. Among others, Ifremer/CERSAT datasets have been used to estimate pollution origin location (H.
Goodwin from the Norwegian Polar Institute, personal comm.), to study sea ice drift (Spreen et al., 2011), and to estimate of sea ice flux in Fram
Strait or Baffin Bay (for example Spreen et al., 2006; Spreen et al., 2009; Kwok, 2007).
They are also commonly used in models, in particular for validation (Martin and Gerdes, 2007 and recently Rozmann et al., 2011). They are also
used in operational models for assimilation. Tests have been done in the UK MetOffice FOAM model showing better realistic results for sea ice
motion when the Ifremer/CERSAT products are assimilated (Stark et al., 2008). Assimilation of these fields greatly improves results on the mod-
elled drift but also on sea ice concentration and thickness for Rollenhagen et al. (2009). Similarly, the assimilation in the TOPAZ systems enables
the modelled sea ice thickness to be more reliable (Bertino et al., 2011).
Conclusions
Satellites provide a unique monitoring capability of sea ice dynamics. Sea ice maps from satellite data have the advantage to cover the whole
polar areas every day. This allows better coverage than that of buoys data, which have a precise daily timing, better spatial resolution but a very
coarse spatial distribution.
Sea ice drifts inferred only from radiometers have a reasonable accuracy but are limited by data gaps and low data density at the beginning and
the end of the cold period. The optimal merging of independent fields of drift data (SSM/I fields and scatterometer fields from QuikSCAT or
ASCAT) at the same resolution improves the data density and the usable time period over winter. It also enables the discrimination of the vector
outliers remaining in the individual products.
The scatterometers (QuikSCAT/ASCAT) drift fields and filled Merged (SSM/I/QuikSCAT and ASCAT/SSMI) Arctic sea ice drift fields datasets are
available every day during winter (October until April). The filled merged fields are available since 1992, using SSM/I horizontal and vertical polari-
zations data from 1992 until 1999 and merging with QuikSCAT data for the 1999-2009 period, ASCAT is merged with SSM/I data since 2007.
Monthly drifts are also computed. The interpolated merged field has the advantage to be gap-filled motion fields ready-to-use.
Ifremer/CERSAT hosts a unique database of 18 years winter time series of sea ice drift. This dataset is available for oceanic and climate model-
ling and various scientific studies in the Arctic. The time series is ongoing and will continue for Arctic long term monitoring using the next MetOp/
ASCAT operational scatterometers, planned to be operated for the next 10 years.
Acknowledgement
The authors thank Robert Ezraty and Julien Paul for their work on the merging and the space and time interpolation developments. They also
thank the CERSAT team for acquisition and archiving of the data and the systematic processing of the sea ice drift maps. SSM/I data are provided
by National Snow and Ice Data Center, University of Colorado (USA), QuikSCAT data by Jet Propulsion Laboratory (USA), and ASCAT raw data
by EUMETSAT. The International Arctic Buoy Programme buoys data are provided by the Polar Science Center, University of Washington (USA).
The Global Monitoring for Environment and Security European Union projects PolarView and MyOcean support partly the data production and
distribution.
References
Aagaard, K., Swift, J. H. and Carmack, E. C. 1985: Thermohaline circulation in the Arctic Mediterranean seas. J. Geophys. Res. 90 (C3): 4833–
4846.
Bertino, L., Sakov, P., and Counillon F. 2011: The TOPAZ sea ice and ocean pilot reanalysis (2003-2008). European Geophysical Union General
Assembly meeting, 3-8 April 2011, Vienna, Austria.
Ifremer/CERSAT ArcƟc sea ice driŌ 1992-2010 Ɵme series from satellite measurements for myocean
25. #43—October 2011—25Mercator Ocean Quarterly Newsletter
Colony, R. and Thorndike, S. 1984: An estimate of the mean field of Arctic sea ice motion. J. Geophys. Res. 89 (C6): 10623–10629.
Emery, W. J., Fowler, C. W., Hawkins, J. and Preller, R. H. 1991: Fram strait satellite image-derived ice motions. J. Geophys. Res. 96 (C3): 4751–
4768.
Emery, W. J., Fowler, C. W. and Maslanik, J. A. 1997: Satellite-derived maps of Arctic and Antarctic sea ice motion: 1988 to 1994. Geophys. Res.
Lett. 24 (8): 897–900.
Ezraty, R., Girard-Ardhuin, F., Piollé, J. F., Kaleschke, L. and Heygster, G. 2007a: Arctic and Antarctic sea ice concentration and Arctic sea ice
drift estimated from Special Sensor Microwave data, User’s manual, version 2.1, February 2007. Available at IFREMER/CERSAT.
Ezraty, R., Girard-Ardhuin, F. and Piollé, J. F. 2007b: Sea ice drift in the central Arctic combining QuikSCAT and SSM/I sea ice drift data, User’s
manual, version 2.0, February 2007. Available at IFREMER/CERSAT.
Ezraty, R., Girard-Ardhuin, F. and Croizé-Fillon, D. 2007c: Sea-ice drift in the central Arctic using the 89 GHz brightness temperatures of the Ad-
vanced Microwave Scanning Radiometer, User’s manual, version 2.0, February 2007. Available at IFREMER/CERSAT.
Girard-Ardhuin, F., Ezraty, R., Croizé-Fillon, D., and Piollé, J.F. 2008: Sea-ice drift in the central Arctic combining QuikSCAT and SSM/I sea ice
drift data, User’s manual, version 3.0, April 2008. Available at IFREMER/CERSAT.
Kamachi, M., 1989: Advective surface velocities derived from sequential images for rotational flow field : limitations and applications of Maximum
Cross Correlation method with rotational registration. J. Geophys. Res. 94 (C12): 18227–18233.
Kwok, R., Curlander, J.C., McConnell, R. and Pang, S. S. 1990: An ice motion tracking system at the Alaska SAR Facility. IEEE. J. Ocean. Eng.
15 (1): 44–54.
Kwok, R., Schweiger, A., Rothrock, D. A., Pang, S. and Kottmeier, C. 1998: Sea ice motion from satellite passive microwave imagery assessed
with ERS SAR and buoy motions. J. Geophys. Res. 103 (C4): 8191–8214.
Kwok, R. 2007: Baffin bay ice drift and export: 2002-2007. Geophys. Res. Lett. 34, doi:10.1029/2007GL031204.
Lavergne, T., Eastwood, S., Teffah, Z., Schyberg, H. and Breivik, L.A. 2010: Sea ice motion from low resolution satellite sensors: an alternative
method and its validation in the Arctic. J. Geophys. Res. 115, doi:10.1029/2009JC005958.
Liu, A. K. and Cavalieri, D. J. 1998: On sea ice drift from the wavelet analysis of the Defense Meteorological Satellite Program (DMSP) Special
Sensor Microwave Imager (SSM/I) data. Int. J. Rem. Sens. 19 (7): 1415–1423.
Liu, A. K., Zhao Y. and Wu, S. Y., 1999: Arctic sea ice drift from wavelet analysis of NSCAT and special sensor microwave imager data. J.
Geophys. Res. 104 (C5): 11529–11538.
Martin, T. and Augstein, E. 2000: Large-scale drift of Arctic sea ice retrieved from passive microwave satellite data. J. Geophys. Res. 105 (C4):
8775–8788.
Martin T., and Gerdes, R. 2007: Sea ice drift variability in Arctic ocean model intercomparison project models and observations. J. Geophys. Res.
112. doi:10.1029/2006JC003617.
Ninnis, R. M., Emery, W. J. and Collins, M. J. 1986: Automated extraction of pack ice motion from Advanced Very High Resolution Radiometer
imagery. J. Geophys. Res. 91 (C9): 10725–10734.
Rollenhagen, K., Timmermann, R., Janjic, T., Schröter, J. and Danilov, S. 2009: Assimilation of sea ice motion in a finite-element sea ice model. J.
Geophys. Res. 114. doi:10.1029/2008JC005067.
Rozmann, P., Hölemann, J., Krumpen, T., Gerdes, R., Köberle, C., Lavergne, T., Adams, S., and Girard-Ardhuin, F. 2011: Validating satellite
derived and modeled sea ice drift in the Laptev sea with In Situ measurements of winter 2007/08. Polar Research. 30 (7218). doi: 10.3402/
polar.v30i0.7218.
Spreen, G., Kern, S., Stammer, D., Forsberg, R. and Haarpaintner, J. 2006: Satellite-based estimates of sea ice volume flux through Fram strait.
Annals of Glaciology. 44: 321-328.
Ifremer/CERSAT ArcƟc sea ice driŌ 1992-2010 Ɵme series from satellite measurements for myocean
26. #43—October 2011—26Mercator Ocean Quarterly Newsletter
Spreen, G., Kern, S., Stammer, D., and Hansen, E. 2009: Fram strait sea ice volume export estimated between 2003 and 2008 from satellite data.
Geophys. Res. Lett. 36. doi:10.1029/2009GL039591.
Spreen, G., Kwok, R., Menemenlis, D. 2011: Trends in Arctic sea ice drift and role of wind forcing: 1992-2009. Geophys. Res. Lett.
doi:10.1029/2011GL048970, in press.
Stark, J., Ridley, J., Martin, M. and Hines, A. 2008: Sea ice concentration and motion assimilation in a sea ice-ocean model. J. Geophys. Res.
113. doi :10.1029/2007JC004224.
Thorndike, A. S. and Colony, R. 1982: Sea ice motion in response to geostrophic winds. J. Geophys. Res. 87: 5845–5852.
Zhao, Y., Liu A. K. and Long, D. G. 2002: Validation of sea ice motion from QuikSCAT with those from SSM/I and buoy. IEEE Trans. On Geosc.
and Rem. Sens. 40 (6): 1241–1246.
Ifremer/CERSAT ArcƟc sea ice driŌ 1992-2010 Ɵme series from satellite measurements for myocean
27. #43—October 2011—27Mercator Ocean Quarterly Newsletter
NOTEBOOK
Notebook
Articles
News: Workshop on Observing System Evaluation and
Intercomparisons : a GODAE Ocean view/GSOP/CLIVAR event (13-
17 June 2011, Santa Cruz, USA)
CORA3, a comprehensive and qualified ocean in-situ dataset from
1990 to 2010
By Cécile Cabanes, Antoine Grouazel, Victor Turpin, François Paris, Christine
Coatanoan, Karina Von Schuckmann, Loic Petit de la Villeon , Thierry Carval ,
Sylvie Pouliquen
A multivariate reanalysis of the North Atlantic ocean
biogeochemistry during 1998-2007 based on the assimilation of
SeaWiFS data
By Clément Fontana, Pierre Brasseur, Jean-Michel Brankart
Ifremer/CERSAT Arctic sea ice drift 1992-2010 time series from
satellite measurements for myocean
By Fanny Girard-Arhuin, Denis Croize-Fillon
Editorial Board
Laurence Crosnier
DTP Operator
Fabrice Messal
Contact :
Please send us your comments to the following e-mail address: webmaster@mercator-ocean.frUH
Next issue : January 2012