CoDICE Lo L3 DE: new variables (Phase D deferred) - Card 3029

[jtniehof]

Summary

The Lo direct events need several new variables, which were discussed briefly at the SWT.

Requirements

• Implement normalization and normalization_per_event variables in l3a direct events. Documented in algorithm document of January 2026, R3C0 (page 58, "normalization factor"). The version on galaxy is up to date even though the document headers say R2C0 and the title page says Oct 2025. See notes in first comment, which supplement the algorithm.
  • Both of these variables should have a floor of 1. Anything less than 1, or 0/0, should be 1.
  • For metadata: these are unitless (units ' '), VALIDMIN 1, VALIDMAX 1000, SCALEMIN 0 (to allow space to see the line at 1), SCALEMAX 10, linear scale, VAR_TYPE data but empty VARIABLE_PURPOSE, lineplot.
• Carry through rgfo_spin_sector, rgfo_esa_step, nso_spin_sector, nso_esa_step from the L2 direct events. Can just copy the metadata as well (either directly from the input L2 CDF, or from the YAML definitions). EDIT: l1a priority, as below probably just use sw input.
• Carry through half_spin_per_esa_step from the l1a priority; it should be the same in both so can just use the sw input. Similar metdata handling to rgfo_spin_sector etc.

This work does not include updates to the 3D VDF code to use the new variables; see #53.

Related tickets

#53

[jtniehof]

This is documented in algorithm document R3C0 pg 58, "normalization factor".

1. L1a DE: sw has P0-P4 variables, nsw has P5-P6. These need to be stacked into a single array.
2. Each spin sector is sampling two opposite directions with each APD; then the priority counts repeat in each half of spin. Trust this :)
3. [:12] and [12:24], i.e. 12 sectors on each.
4. no notes
5. direct events report spin sectors 0-23. Loop n for 0 through 11, and then the second half is picked up with the second term. (Again because the APDs sample opposite directions)
6. this is filling the back half like step 3, again 12:24 and 0:12. Basically this is direct event histogramming. (There is only one p in the indexing).
7. no notes
8. Both normalization and normalization_per_event need to be added to the L3a direct events.

[jtniehof]

Also need to carry through rgfo_spin_sector, rgfo_esa_step, nso_spin_sector, nso_esa_step from the L2 direct events. All we can probably just copy metadata from L2. Also need to copy over half_spin_per_esa_step from the l1a priority; it should be the same in both so can just use the sw.