Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
SlideShare a Scribd company logo

Rehder, Gregor: Making Atmopheric measurements onboard SOOPs

•Download as PPTX, PDF•
•
Integrated Carbon Observation System (ICOS)
Integrated Carbon Observation System (ICOS)

This document discusses improving atmospheric measurements on Ships of Opportunity (SOOP) as part of the Ringo Task 3.2 project. Three SOOP lines - SOOP Tavastland, SOOP Colibri, and SOOP Atlantic Sail - are being equipped with instrumentation to measure atmospheric CO2, CH4, and other greenhouse gases to standards matching the ICOS Atmospheric Thematic Centre. Initial results from SOOP Colibri show data quality matching these standards and capturing gradients and variability not seen in models. Next steps include finalizing the setup on SOOP Tavastland and assessing the added value of these new SOOP measurements for inverse modeling.

1 of 13
High quality trace gas measurements in the Marine Boundary Layer on Ships of Opportunity (SOOP) “Ringo Task 3.2” Gregor Rehder, Marc F. Delmotte, Michael Glockzin, Lynn Hazan, Armin Jordan, Victor Kazan, Syrine Nouicer, Michel Ramonet, and Tobias Steinhoff gregor-rehder@io-warnemuende.de
BACKGROUND AND RATIONALE - A fleet of SOOP lines hosting automated seawater pCO2 measurements provides backbone for global pCO2 integration (e.g. SOCAT) - Most SOOP lines maintain atm. measurements (~every 6-8 h) in the MBL - Use of atm. measurements mostly limited to QC purposes, or local ASE calculations - Atm. data from SOOPs have so far not been regularly quality controlled or transferred into products - Yet, if quality was sufficient, SOOP lines could provide atm. CO2 and other GHG concentrations at locations difficult to reach, on platforms which are maintained anyhow - Data with “exotic” spatiotemporal resolution from Wanninkhof et al., 2019 from Steinhoff et al., 2019
RINGO TASK 3.2. Improving atmospheric measurements on Ships Of Opportunity (SOOP) Overarching Questions: - Can we provide data matching ICOS ATC criteria from SOOPs, using the same technology? - How far can we improve sporadic atm. measurements using existing pCO2 setups? - Would these data improve eg. inverse modelling products? - Could data be handled by the ATC? THREE COMLEMENTARY LINES SOOP TAVASTLAND (IOW) Baltic (Lübeck – Kemi) - easy to access, - surrounded by ICOS atm. station network - aim for „easy to move“-unit - covering continental to Arctic air masses SOOP COLIBRI (LSCE) Europe – French Guiana - undersampled equatorial Atlantic - extreme range of external conditions SOOP ATLANTIC SAIL (GEOMAR) English Channel – Westwards - attempt to use instrumentation also used for water measurements - cost optimization (MPI JENA, CAL, and ICOS ERIC) Consulting during construction phase, provision of flask sampler, provision of calibration gases, assessment of added value through inverse modelling runs
Current status • Severe delays due to instrument availability, ship liaison issues, and lately ship access due to Covid 19 • Yet we are getting there • Glimpse on results with strong focus on SOOP COLIBRI International Activity - Recognition of the potential of SOOP-hosted MBL measurements in Ocean Obs 2019 Community paper (Wanninkhof et al., 2019) - Proof of concept underway using data from the UK Carribean Line (UExeter, UEA, UEdingburgh) against GEOS-Chem atmospheric chemistry transport model
North Atlantic Line (SOOP ATLANTIC SAIL, GEOMAR) •Installation of suite of calibration gases and target gas ranging from 373 to 448 ppm (provided by ICOS CAL • Air filtering system and add. drying step • Parallel run of LICOR 7000 (classical infrared) and new LICOR 7815 (OF-CEAS) Results and Importance - Tests suggest repeatability of better 0.01 ppm at tav = 10 seconds for the LI 7815 - Add. tests on long-term drift needed - Required modifications on the GO system due to internal pump of the sensor - Integration in GO system internationally on the way; likely to become standard set-up
Picarro pump Picarro Analyser Multi-positions Valve Safeguard hard disk UPS o Air inlet o CRDS analyser (CO2, CH4, CO, H2O - Picarro G2401) o Calibration and quality control compressed air cylinders (4) provided by the central calibration centre from ICOS-RI connected to a multi-position valve o GPS set up to recover real time positioning of the vessel and absolute time stamp o Power supply to preserve the equipment in case of power shut-down o 4G router used for data transfer during the time the vessel stays in a harbor GPS Air inlet line and GPS cable Installation on SOOP COLIBRI Inlet
Data • Regular Target Measurements March- Sept. 2019 with biases lower than 0.05 ppm for CO2 and 0.5 ppb for CH4 • Achievable data quality matches ATC standards • Data stream from moving platform was implemented • Different treatment in harbor, coastal, and open ocean settings by automated data selection (based on position information)
Data • Regular Target Measurements March-Sept 2019 with biases lower than 0.05 ppm for CO2 and 0.5 ppb for CH4 • Achievable data quality matches ATC standards • Data stream from moving platform was implemented • Different treatment in harbor, coastal, and open ocean settings by automated data selection (based on position information) • Spike detection implemented (after El Yazidi et al. 2018); CO contamination as contamination proxy Concentration differences with and without spikes CO Spikes are found in hours for 27% in harbour, 9% in European coasts and 6% during transtalantic passsages Tableau3.1 : Valeurs moyennes de l’écartentre lesmesures CO2, CH4 et CO avant et après l’éliminatio pics de CO Port Europe Atlantique CO2 (ppm) 1.2 0.7 0.5 CH4 (ppb) 3.3 0.15 -0.002 CO (ppb) 1.4 0.6 0.7 Mean differences of hourly means with and without spikes
So far 5 round trips Europe > Guyana > Europe, each one lasting about 1 month
Strong correlations betweens the 3 species North/South gradient: increase of CO and CH4 at high latitudes, and decrease of CO2 (biospheric uptake) CO & CO2 higher near Amazonia, maybe due to the strong fire activity in Aug. 2019 Short term variability explained by synoptic transport of air masses. Aug-Sept 2019
Comparison of CO2 and CH4 with CAMS forecasts • Systematic offset of CO2 and CH4 concentrations. These offset are already well known and documented in reports based on ICOS and TCCON observations • Most synoptic variabilities are very well represented in location and amplitude • The signals from Amazonia (CO2 increase, CH4 decrease) are underestimated Aug-Sept 2019
NEXT STEPS • Final assembly of instrumentation on board SOOP Tavastland (all components on ship, but assembly only with one person during the 6h harbor window on Fridays) • Analysis of the first months of atm. measurements using “just” short atm. data from a pCO2 system equipped with a LI 7815 (SOOP ATLANTIC SAIL) • Assessment of these data by ATC • Analysis of improvement for e.g inverse modelling products
High quality trace gas measurements in the Marine Boundary Layer on Ships of Opportunity (SOOP) “Ringo Task 3.2” Gregor Rehder, Marc F. Delmotte, Michael Glockzin, Lynn Hazan, Armin Jordan, Victor Kazan, Syrine Nouicer, Michel Ramonet, and Tobias Steinhoff gregor-rehder@io-warnemuende.de Thank you Questions?

More Related Content

Rehder, Gregor: Making Atmopheric measurements onboard SOOPs

  • 1. High quality trace gas measurements in the Marine Boundary Layer on Ships of Opportunity (SOOP) “Ringo Task 3.2” Gregor Rehder, Marc F. Delmotte, Michael Glockzin, Lynn Hazan, Armin Jordan, Victor Kazan, Syrine Nouicer, Michel Ramonet, and Tobias Steinhoff gregor-rehder@io-warnemuende.de
  • 2. BACKGROUND AND RATIONALE - A fleet of SOOP lines hosting automated seawater pCO2 measurements provides backbone for global pCO2 integration (e.g. SOCAT) - Most SOOP lines maintain atm. measurements (~every 6-8 h) in the MBL - Use of atm. measurements mostly limited to QC purposes, or local ASE calculations - Atm. data from SOOPs have so far not been regularly quality controlled or transferred into products - Yet, if quality was sufficient, SOOP lines could provide atm. CO2 and other GHG concentrations at locations difficult to reach, on platforms which are maintained anyhow - Data with “exotic” spatiotemporal resolution from Wanninkhof et al., 2019 from Steinhoff et al., 2019
  • 3. RINGO TASK 3.2. Improving atmospheric measurements on Ships Of Opportunity (SOOP) Overarching Questions: - Can we provide data matching ICOS ATC criteria from SOOPs, using the same technology? - How far can we improve sporadic atm. measurements using existing pCO2 setups? - Would these data improve eg. inverse modelling products? - Could data be handled by the ATC? THREE COMLEMENTARY LINES SOOP TAVASTLAND (IOW) Baltic (Lübeck – Kemi) - easy to access, - surrounded by ICOS atm. station network - aim for „easy to move“-unit - covering continental to Arctic air masses SOOP COLIBRI (LSCE) Europe – French Guiana - undersampled equatorial Atlantic - extreme range of external conditions SOOP ATLANTIC SAIL (GEOMAR) English Channel – Westwards - attempt to use instrumentation also used for water measurements - cost optimization (MPI JENA, CAL, and ICOS ERIC) Consulting during construction phase, provision of flask sampler, provision of calibration gases, assessment of added value through inverse modelling runs
  • 4. Current status • Severe delays due to instrument availability, ship liaison issues, and lately ship access due to Covid 19 • Yet we are getting there • Glimpse on results with strong focus on SOOP COLIBRI International Activity - Recognition of the potential of SOOP-hosted MBL measurements in Ocean Obs 2019 Community paper (Wanninkhof et al., 2019) - Proof of concept underway using data from the UK Carribean Line (UExeter, UEA, UEdingburgh) against GEOS-Chem atmospheric chemistry transport model
  • 5. North Atlantic Line (SOOP ATLANTIC SAIL, GEOMAR) •Installation of suite of calibration gases and target gas ranging from 373 to 448 ppm (provided by ICOS CAL • Air filtering system and add. drying step • Parallel run of LICOR 7000 (classical infrared) and new LICOR 7815 (OF-CEAS) Results and Importance - Tests suggest repeatability of better 0.01 ppm at tav = 10 seconds for the LI 7815 - Add. tests on long-term drift needed - Required modifications on the GO system due to internal pump of the sensor - Integration in GO system internationally on the way; likely to become standard set-up
  • 6. Picarro pump Picarro Analyser Multi-positions Valve Safeguard hard disk UPS o Air inlet o CRDS analyser (CO2, CH4, CO, H2O - Picarro G2401) o Calibration and quality control compressed air cylinders (4) provided by the central calibration centre from ICOS-RI connected to a multi-position valve o GPS set up to recover real time positioning of the vessel and absolute time stamp o Power supply to preserve the equipment in case of power shut-down o 4G router used for data transfer during the time the vessel stays in a harbor GPS Air inlet line and GPS cable Installation on SOOP COLIBRI Inlet
  • 7. Data • Regular Target Measurements March- Sept. 2019 with biases lower than 0.05 ppm for CO2 and 0.5 ppb for CH4 • Achievable data quality matches ATC standards • Data stream from moving platform was implemented • Different treatment in harbor, coastal, and open ocean settings by automated data selection (based on position information)
  • 8. Data • Regular Target Measurements March-Sept 2019 with biases lower than 0.05 ppm for CO2 and 0.5 ppb for CH4 • Achievable data quality matches ATC standards • Data stream from moving platform was implemented • Different treatment in harbor, coastal, and open ocean settings by automated data selection (based on position information) • Spike detection implemented (after El Yazidi et al. 2018); CO contamination as contamination proxy Concentration differences with and without spikes CO Spikes are found in hours for 27% in harbour, 9% in European coasts and 6% during transtalantic passsages Tableau3.1 : Valeurs moyennes de l’écartentre lesmesures CO2, CH4 et CO avant et après l’éliminatio pics de CO Port Europe Atlantique CO2 (ppm) 1.2 0.7 0.5 CH4 (ppb) 3.3 0.15 -0.002 CO (ppb) 1.4 0.6 0.7 Mean differences of hourly means with and without spikes
  • 9. So far 5 round trips Europe > Guyana > Europe, each one lasting about 1 month
  • 10. Strong correlations betweens the 3 species North/South gradient: increase of CO and CH4 at high latitudes, and decrease of CO2 (biospheric uptake) CO & CO2 higher near Amazonia, maybe due to the strong fire activity in Aug. 2019 Short term variability explained by synoptic transport of air masses. Aug-Sept 2019
  • 11. Comparison of CO2 and CH4 with CAMS forecasts • Systematic offset of CO2 and CH4 concentrations. These offset are already well known and documented in reports based on ICOS and TCCON observations • Most synoptic variabilities are very well represented in location and amplitude • The signals from Amazonia (CO2 increase, CH4 decrease) are underestimated Aug-Sept 2019
  • 12. NEXT STEPS • Final assembly of instrumentation on board SOOP Tavastland (all components on ship, but assembly only with one person during the 6h harbor window on Fridays) • Analysis of the first months of atm. measurements using “just” short atm. data from a pCO2 system equipped with a LI 7815 (SOOP ATLANTIC SAIL) • Assessment of these data by ATC • Analysis of improvement for e.g inverse modelling products
  • 13. High quality trace gas measurements in the Marine Boundary Layer on Ships of Opportunity (SOOP) “Ringo Task 3.2” Gregor Rehder, Marc F. Delmotte, Michael Glockzin, Lynn Hazan, Armin Jordan, Victor Kazan, Syrine Nouicer, Michel Ramonet, and Tobias Steinhoff gregor-rehder@io-warnemuende.de Thank you Questions?