Settle the hierarchy of variables and attributes for Leslie, and fix a new json format

[sebastien-lenard]

Aim: have a view of how variables organize in a 'tree'. This is for helping Leslie and us, but we could have no tree at all in the website for development simplicity and rather decide of a big list of available variables.

Additionally, the current json format of variables should be updated. It might be necessary to split in two:

• 1 file for global parameters, such as colors
• 1 file for each root region, to indicate which variable is available for that root region.

---

First draft of tree:

sensor  platform    algorithm   variable name               variable type   rank in list

modis
        |_ Terra
                  |_ stc    
                             |_ snow cover %                absolute        1
                             |_ radiative forcing           absolute        2
                             |_ albedo                      absolute        3
                             |_ snow cover days             absolute        4
                             |_ snow cover anomaly          anomaly         5
                             |_ radiative forcing anomaly          anomaly         6
                             |_ albedo anomaly              anomaly         7
                             |_ snow cover days anomaly     anomaly         8

                  |_ spires
                             |_ snow cover %                absolute        11
                             |_ radiative forcing           absolute        12
                             |_ albedo                      absolute        13
                             |_ snow cover days             absolute        14
                             |_ snow cover anomaly          anomaly         15
                             |_ radiative forcing anomaly          anomaly         16
                             |_ albedo anomaly              anomaly         17
                             |_ snow cover days anomaly     anomaly         18

viirs
        |_ ?
                  |_ stc    
                             |_ snow cover %                absolute        21
                             |_ radiative forcing           absolute        22
                             |_ albedo                      absolute        23
                             |_ snow cover days             absolute        24
                             |_ snow cover anomaly          anomaly         25
                             |_ radiative forcing anomaly          anomaly         26
                             |_ albedo anomaly              anomaly         27
                             |_ snow cover days anomaly     anomaly         28

                  |_ spires
                             |_ snow cover %                absolute        31
                             |_ radiative forcing           absolute        32
                             |_ albedo                      absolute        33
                             |_ snow cover days             absolute        34
                             |_ snow cover anomaly          anomaly         35
                             |_ radiative forcing anomaly          anomaly         36
                             |_ albedo anomaly              anomaly         37
                             |_ snow cover days anomaly     anomaly         38

Remarks:

• The tree structure remains the same whatever the root region
• But for some root regions: data for a sensor, or an algorithm, or for some variables might be unavailable.
• The max number of sensors /platform / algorithm will probably remains the same for the next 2 years.

Other Remarks for potential info to display in the website:

• There is this additional issue that for some root regions, we have the historics and for others we don't have, or we have, but only for 5 years instead of 23 years. This info should also be somewhere.
• Additionally, for some root regions, the historics will be originate from a source called JPL and for others the historics will originate from the DAAC. This info should also be available

[sebastien-lenard]

@MattF-NSIDC when you'll have a bit of time would you mind thinking about what could be the update of the json structure for variables?

[MattF-NSIDC]

@windnagelnsidc don't miss this part:

Other Remarks for potential info to display in the website:

• There is this additional issue that for some root regions, we have the historics and for others we don't have, or we have, but only for 5 years instead of 23 years. This info should also be somewhere.
• Additionally, for some root regions, the historics will be originate from a source called JPL and for others the historics will originate from the DAAC. This info should also be available

[windnagelnsidc]

@MattF-NSIDC Thanks for drawing my attention to that. I'll make sure to bring it up on Tuesday.

[MattF-NSIDC]

Any time :)

Is it appropriate to think of these variables as a 2d matrix of independent variables? "Feature measured" along the top, and "statistic" along the left?

	snow cover %	radiative forcing	albedo
absolute
anomaly (clim)
anomaly (annual)

EDIT: I see now that "snow cover days" does not have an anomaly calculation. Is that intentional? I guess it makes sense that this wouldn't have an anomaly value because it's a cumulative statistic? So the 2d matrix view falls apart. Reply: @MattF-NSIDC It's a mistake, all variables have anomalies. But we still may decide to remove the display of one anomaly but keeping the absolute value, in case of whatever problem

[windnagelnsidc]

I'm going to add this as a topic to ask Karl for the Tuesday stake holder meeting.