Naming of W variables.

[eleanorfrajka]

Figured I'd open an issue to discuss how to name vertical velocity variables in the OG1 format.

For the calculation of vertical seawater velocity using e.g., Frajka-Williams et al. 2011 or Merckelbach et al. 2010, we need 2 things:

• Modelled glider vertical velocity ($w_{stdy}$ in Frajka-Williams and $w_g$ in Merckelbach)
• Calculated glider vertical velocity from pressure ($w_{meas}$ in Frajka-Williams and $w_p$ in Merckelbach).

These are equations (1) in Frajka-Williams: ($w_w = w_{meas}-w_{stdy}$), or equivalently equation (15) in Merckelbach, ($w = w_p - w_g$).

Example: if the glider is diving, the w_meas and w_stdy will be something like -10 cm/s (-0.1 m/s). But suppose the measured vertical velocity (from dz/dt) is more negative, e.g., it's in a convective plume and the measured vertical velocity is -15 cm/s. The modelled vertical velocity mainly changes due to glider density or seawater density changes so is still -10 cm/s. Then we would have
w_w = w_meas - w_stdy = (-15 cm/s) - (-10 cm/s) =-5 cm/s

Proposed naming:

• Could use W to indicate vertical speed, but there appears to be some things like VERT_SPEED already
• Could use VERT_SPEED but then need something to indicate the source of the measurement, and noting that possibly something like this could come from other sensors. But it's not really a "SPEED" in the sense that WIND_SPEED is, since WIND_SPEED also comes with WIND_DIRECTION, and this won't.
• Could use VERT_VELO
  Any future conflicts with a "w" or "vertical speed" parameter that could come from another sensor?

Long names:

• VERT_SPEED_GLIDER_MODEL: "modelled vertical glider speed"
• VERT_SW_SPEED: "vertical seawater speed"
• VERT_SPEED_DZDT: "vertical seawater speed from pressure measurements"

Conventions on units:

• Use SI as standard (so, m/s?) or "units": "m s-1"
• Use cm/s or "units": "cm s-1",

Other attributes:
"positive": "up"
"observation_type": "calculated",

• And in the future, might be useful to be able to pass details on VERT_SPEED_GLIDER_MODEL like what model was used (there are at least 2 for Seaglider and 2 for Slocum).

For working with w, any value in defining a new depth Z which is "positive" "up". This would be consistent with the Gibbs Seawater toolbox calculation of depth from pressure. And would ensure that people don't accidentally use DEPTH which would give a sign error. Or, I guess I could figure out how to work with these "positive" "up" attributes.