Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.
description
Transcript of Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.
![Page 1: Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.](https://reader036.fdocuments.us/reader036/viewer/2022070410/5681459a550346895db28cca/html5/thumbnails/1.jpg)
Progress towards a 16-month CS510 adjoint state estimate
Hong Zhang, Dimitris Menemenlis
JPL/Caltech
Gael Forget, Patrick Heimbach, Chris Hill
MIT/EAPS
ECCO2 meeting, Pasadena, 2009
![Page 2: Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.](https://reader036.fdocuments.us/reader036/viewer/2022070410/5681459a550346895db28cca/html5/thumbnails/2.jpg)
ObjectiveObtain an eddying global ocean and sea ice state estimate using the adjoint method.
Outline1. Introduction2. Results
a. overall cost reductionb. IC, BC adjustmentsc. RMS model-data difference
3. Concluding Remarks
![Page 3: Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.](https://reader036.fdocuments.us/reader036/viewer/2022070410/5681459a550346895db28cca/html5/thumbnails/3.jpg)
One year ago, from Gael’s presentation at the MIT ECCO2
meeting
![Page 4: Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.](https://reader036.fdocuments.us/reader036/viewer/2022070410/5681459a550346895db28cca/html5/thumbnails/4.jpg)
One year ago, from Patrick’s presentation at the MIT
ECCO2 meeting
![Page 5: Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.](https://reader036.fdocuments.us/reader036/viewer/2022070410/5681459a550346895db28cca/html5/thumbnails/5.jpg)
Configuration details
● CS510 cube sphere with ~18 km horizontal grid spacing and 50 vertical levels
● First guess initial conditions from a 2-yr spin up from OCCA climatology
● First guess atmospheric state from ECMWF
● Constraints: OCCA climatology, ARGO, XBT, Jason, Envisat, and AMSR-E
● Currently optimizing 16 months starting January 2004, i.e., the beginning of the ARGO period
● Using 900 cpus on Columbia or 3600 cpus on Pleiades (for extra memory to reduce adjoint re-computations)
![Page 6: Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.](https://reader036.fdocuments.us/reader036/viewer/2022070410/5681459a550346895db28cca/html5/thumbnails/6.jpg)
Overall cost function reduction
![Page 7: Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.](https://reader036.fdocuments.us/reader036/viewer/2022070410/5681459a550346895db28cca/html5/thumbnails/7.jpg)
ARGO (T/S) cost function reduction
![Page 8: Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.](https://reader036.fdocuments.us/reader036/viewer/2022070410/5681459a550346895db28cca/html5/thumbnails/8.jpg)
Adjustments of initial temperature
At 154m
At 634m
![Page 9: Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.](https://reader036.fdocuments.us/reader036/viewer/2022070410/5681459a550346895db28cca/html5/thumbnails/9.jpg)
Adjustments of initial salinity
At 154m
At 634m
![Page 10: Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.](https://reader036.fdocuments.us/reader036/viewer/2022070410/5681459a550346895db28cca/html5/thumbnails/10.jpg)
Adjustments of atmospheric zonal wind
mean
std
![Page 11: Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.](https://reader036.fdocuments.us/reader036/viewer/2022070410/5681459a550346895db28cca/html5/thumbnails/11.jpg)
Adjustments of atmospheric meridional wind
mean
std
![Page 12: Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.](https://reader036.fdocuments.us/reader036/viewer/2022070410/5681459a550346895db28cca/html5/thumbnails/12.jpg)
rms(iter14 – Argo T) – rms(iter0 – Argo T)
At 154m
At 634m
![Page 13: Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.](https://reader036.fdocuments.us/reader036/viewer/2022070410/5681459a550346895db28cca/html5/thumbnails/13.jpg)
At 154m
At 634m
rms(iter14 – Argo S) – rms(iter0 – Argo S)
![Page 14: Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.](https://reader036.fdocuments.us/reader036/viewer/2022070410/5681459a550346895db28cca/html5/thumbnails/14.jpg)
T
At 154m
At 634m
rms(iter14 – OCCA T) – rms(iter0 – OCCA T)
![Page 15: Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.](https://reader036.fdocuments.us/reader036/viewer/2022070410/5681459a550346895db28cca/html5/thumbnails/15.jpg)
At 154m
At 634m
rms(iter14 – OCCA S) – rms(iter0 – OCCA S)
![Page 16: Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.](https://reader036.fdocuments.us/reader036/viewer/2022070410/5681459a550346895db28cca/html5/thumbnails/16.jpg)
rms(iter14 – EVISAT SSH) – rms(iter0 – EVISAT SSH)
rms(iter14 – AMSRE SST) – rms(iter0 – AMSRE SST)
![Page 17: Objective Obtain an eddying global ocean and sea ice state estimate using the adjoint method.](https://reader036.fdocuments.us/reader036/viewer/2022070410/5681459a550346895db28cca/html5/thumbnails/17.jpg)
• A global, eddying, full-depth-ocean and sea-ice configuration of the MITgcm is being constrained with a variety of satellite and in-situ data products using the adjoint method
• After 14 forward-adjoint iterations there is a 35% overall cost function reduction
• TO DO:1. additional data constraints (mean dynamic topography,
QuikSCAT wind stress, sea-ice observations)2. additional controls (e.g., vertical diffusivity)3. improved minimization algorithms (smoothed gradients,
improved line search, etc.)4. longer optimization period5. first science applications …
Concluding Remarks