Argo Paper Tasks #17
Description
Remaining tasks for the Argo project:
- Select pmean_smooth by 3_deg x 3_deg bin for coordinate transformation.
- Gappiness introduced by transformation to pmean space. Why is so much data lost?
- In plots, mask regions where mean lateral gradients go to zero
- Grid spacing used for coordinate transformations: how does the density step relate to pressure? is there sufficient resolution in this transformation, or am I unintentionally smoothing the profiles?
- Lit review: how do my estimates compare to previously published papers? Focus on global studies, where possible.
- What's the appropriate scale for gradient calculations and smoothing?
- Vertical derivatives are computed on 2m grids, smoothed at 100m.
- Horizontal derivates are computed on 3_deg bins
- Smoothing method: length scale filter vs polynomial fit. What filter length scale is most appropriate? (how does changing the filter scale change the variance production, diffusivity estimates?)
- Plot N^2 globally to estimate stratification. (plot based on mean profiles)