Assessment of gas transfer velocities computed from the COARE gas transfer model using satellite-derived inputs

Gary A Wick, NOAA, gary.a.wick@noaa.gov (Presenter)
Darren L Jackson, CIRES/NOAA, darren.l.jackson@noaa.gov

Understanding the processes and rate of exchange of gases between the atmosphere and global oceans is of vital importance to the climate community. While in situ observations from research vessels provide the most accurate observations of air-sea gas exchange, ultimately remote sensing techniques are needed to provide the necessary temporal and spatial sampling to capture seasonal and interannual variations of gas exchange for all oceanic regions. Satellite methods, however, remain challenging due to difficulties in parameterizing the gas/chemical transfer processes and accurately determining and retrieving the essential satellite-derived inputs. The goal of this study is to assess the accuracy of the carbon dioxide (CO2) transfer velocity estimates computed from the physically-based COARE gas transfer model using satellite-derived inputs and to construct a global set of observations and corresponding uncertainty estimates of the transfer velocity.

Several satellite inputs are used for the COARE model including sea surface temperature, downward solar and infrared radiation, and near-surface wind speed, humidity (Qa), and temperature (Ta). Recent updates to the multi-sensor, satellite-based retrievals of Ta and Qa include higher spatial resolution of 0.25 degrees, an improved limb correction technique for AMSU-A, and inclusion of multiple AMSU-A satellites. The satellite inputs were used to derive transfer velocities and compared with ship-derived COARE transfer velocities and direct covariance observations from the recent GasEx cruises. Results indicate good agreement between the satellite-derived CO2 COARE transfer velocities and ship-derived transfer velocities with a bias of less than 1 cm/hr and RMS differences of approximately 7 cm/hr. Satellite-derived NOAA-COARE transfer velocities were found to have a bias of 6.32 cm/hr and RMS difference of 14.02 cm/hr when compared with covariance-derived transfer velocities. These results imply that the model can be applied with satellite inputs with accuracy comparable to that achieved with ship-based observations.

Presentation Type:  Poster

Session:  Coupled Processes at Land-Atmosphere-Ocean Interfaces   (Mon 4:00 PM)

Associated Project(s): 

• Wick, Gary: Assessment and Advancement of Satellite-Based Remote Sensing of Gas Fluxes for the Southern Ocean Air-Sea CO2 Study ...details

Poster Location ID: 96