NOAA

Geophysical Fluid
Dynamics Laboratory

Skip to: [content] [navigation]
search gfdl:

If you are using Navigator 4.x or Internet Explorer 4.x, this site will not render correctly!

gfdl's home page > data portal > deccen coupled climate models > CM2.x coupled climate models > CM2.X model ocean and sea ice grids information

CM2.X model ocean and sea ice grids information


[NOAA bullet] CM2.X model ocean and sea ice grids information.

[Sample image of the tripolar grid]

Tripolar & Interpolated Horizontal Grid Issues:


Some of the GFDL CM2.X ocean and sea ice model output variables are available for download on two different grids: files using the models' native tripolar grid -or- files using a more traditional longitude-latitude Cartesian grid to which the variables have been interpolated. (After interpolation, all points along a given J-row lie at the same latitude, and all points sharing the same I-index lie along the same longitude.) Other variables are available only on the tripolar grid.

Whether an analyst chooses to work with data on the tripolar grid or on the interpolated grid, care must be taken to assure that quantitative analyses and mapping of the model output are done properly. This web page contains information and links related to issues associated with the grids found in the CM2.X ocean and sea ice output files on this server.

NOTE:
South of 65N the ocean and sea ice models' native tripolar grids follow a traditional latitude by longitude arrangement that can be easily ingested by most analysis software, and thus does not require interpolation. Thus, the grids and data values found in the tripolar and interpolated grid files are identical south of 65N. The interpolation process only affects values north of 65N. Also, because the grid dimensions are identical in the tripolar and interpolated grids (360x200x50), the tripolar and interpolated model output files are very similar in size.

Why the tripolar grid?

To avoid a singularity at the North Pole when running the model, the CM2.X ocean and sea ice model components use a grid that has "poles" located in the land masses of northern Canada and northern Russia, in addition to one at the South Pole (one view shown in the figure on this page). There is no artificial land mass inserted at the geographic North Pole, which is something that usually is done for numerical reasons if a spherical coordinate grid is used in the Arctic. One can read more about the tripolar grid in A Technical Guide to MOM 4. In that document, go to section 4.2 titled "The Murray (1996) Tripolar grid."

Though beneficial numerically when running the model, many researchers find the tripolar grid difficult to work with when analyzing model output if their analysis domain includes the Arctic Ocean. Those who wish to analyze or visualize the ocean and sea ice model output on the models' native tripolar grid should read the information on the web page: Analyzing model output on the tripolar grid

Why the interpolated grid?

Analysts can find working with model output on the tripolar grid to be quite challenging. Additionally, GFDL is participating in the IPCC AR4 process, and the IPCC/PCMDI archive requirements mandate that files to be served at their site be presented on a longitude-latitude Cartesian grid. For these reasons, GFDL has worked to develop a conservative interpolation scheme and supplementary files that together will allow one to perform quantitatively accurate calculations with interpolated CM2.X ocean and sea ice model output.

Currently, not all ocean and sea ice variables are available in an interpolated form. The first interpolated files to be made available for download are scalar 2D variables. Next will be scalar 3D variables. Later, we hope to make interpolated 2D and 3D vector quantities available.

Ultimately, the decision of whether we have created a interpolated product suitable for distribution is based upon one simple question: have we produced interpolated variables of high enough quality that we at GFDL would use them for the usual range of quantitative analyses we ourselves perform? If the answer is "Yes" for a particular variable, we will place the interpolated files on this server as we produce them. If the answer is "No", one will need to rely on the tripolar version of the model output for that variable.

While generally easier to work with than the tripolar grid, model output on the interpolated grid presents its own challenges for visualization and to perform quantitatively accurate computations north of 65N. For example, one must be aware that the surface area represented by an individual cell on the interpolated grid may not be all ocean or all land, but rather be partially land and partially ocean. For this reason, we provide supplementary files documenting the surface areas and volumes represented by each of the grid cells. Those who wish to use CM2.X ocean and sea ice model output on the interpolated grid should read the information on the web page: Analyzing model output on the interpolated grid

Background info on the tripolar and interpolated grids:
The GFDL Data Portal & PCMDI AR4 Archives

To be accepted at the IPCC WG1 data archive, it is required for model output files to employ a Cartesian latitide longitude grid. However, the CM2.X ocean and sea ice models use a tripolar grid that does not fit the PCMDI criterion north of 65N. So, to meet the IPCC requirements, interpolation is required for CM2.X ocean and sea ice variables in the Arctic.

At GFDL, we were unable to develop an interpolation scheme that met GFDL quality control standards for all ocean and sea ice variables (scalar and vector, 2D and 3D) in time for the 1 May 2005 deadline for IPCC A4 analysts to have their manuscripts submitted for publication. So, in early 2005, the decision was made to use GFDL's Data Portal to serve the ocean and sea ice model output on the model's native tripolar grid. We also provide users with additional variables and files beyond those requested for the IPCC/PCMDI archive, to aid researchers in their use of the CM2.X ocean and sea ice model output files. This way, quantitatively accurate computations could be made using the tripolar grid.

Now, in 2006, we have developed a scheme that we are using to interpolate some of the ocean and sea ice variables onto a regular grid north of 65N in a manner that meets GFDL quality standards. Throughout 2006 we expect interpolated versions of many CM2 ocean and sea ice variables to be added to both the GFDL Data Portal and at PCMDI. South of 65N the tripolar and interpolated data files employ the same grid and contain the same variable values. These interpolated products will be served on this GFDL Data Portal and shipped to PCMDI.

Questions related to the GFDL CM2.x models may be directed to…


[email GFDL.Climate.Model.Info at noaa dot gov]