T3 — the null census of AR11158 through its emergence (2011-02-13 → 02-15)
Reproduce: ../cosmic-web/.venv/bin/python scripts/fetch_sequence.py (8 real HMI frames,
~125 MB, gitignored) then scripts/run_p5_sequence.py (~6 s). Results →
artifacts/p5_sequence.json, figure → public/img/posts/forbidden-directions-emergence.png.
What was hunted, and what was caught
The goal: catch a fold — an opposite-sign null pair being born — on the real Sun, and watch the S1 knee migrate. The target: AR11158, the textbook emergence region (born Feb 12–13, X2.2 flare Feb 15 01:56), tracked across 2.5 days at 6–12 h cadence, window locked on the flux centroid (the first, uncentred tracking pass produced a spurious “annihilation candidate” from quiet-Sun framing — caught and discarded).
The census (interior coronal nulls of the potential extrapolation):
| frame | 13T00 | 13T06 | 13T12 | 13T18 | 14T00 | 14T12 | 15T00 | 15T12 |
|---|---|---|---|---|---|---|---|---|
| nulls | 0 | 0 | 0 | 1* | 0 | 1 | 0 | 2 |
| peak G | 1379 | 1578 | 1743 | 1713 | 1833 | 1884 | 1749 | 1726 |
(*threshold-level transient.)
Honest verdict
- The physics trend is captured: a young emerging bipole is a simple arcade — zero coronal nulls through day one — and nulls appear as the region builds its multipolar (flare-capable) topology; by twelve hours after the X2.2 flare the volume hosts a null pair (separation 36.5 px).
- A fold birth was NOT caught. The final pair is radial+/radial+ — same sign — and fold-born pairs are necessarily opposite-signed (topological degree). These two nulls arose independently (or with partners outside the box / below threshold).
- What a real catch needs (recorded): hourly cadence, a co-moving window locked to the same photospheric flux (not re-framed per frame), and null identity tracking between frames — plus the honest caveat that a potential extrapolation of a flaring (highly non-potential) region approximates the topology at best.
Status
T3’s minimum deliverable (the census time series) is delivered, with one genuinely useful real-data observation — null count as a topological complexity index of an emerging region — and the fold hunt honestly unresolved, with its data requirements specified.