NILC and SEVEM: Needlet Localisation and Template-Fitting Residual Budgets Compared
NILC, which minimises variance locally in needlet space, and SEVEM, which subtracts foreground templates built from channel differences, reach similar high-latitude accuracies by opposite philosophies: NILC leaves residuals near 1.8 microkelvin at degree scales by adapting to local foreground conditions, while SEVEM's residuals are set by template noise and coefficient errors near a comparable level with different spatial character. The budget tabulates adopted NILC residuals at three angular scales, computes SEVEM residual contributions from template noise propagated through stated coefficients, forms the expected root-sum-square difference between the two methods' background maps, and compares that expectation with the published inter-method difference to test whether the two error models are mutually consistent. Headline values are nilcsev nilc residual large uk nilcsev = 2.6; nilcsev nilc residual degree uk nilcsev = 1.9; nilcsev nilc residual small uk nilcsev = 1; nilcsev sevem template1 contribution uk nilcsev = 0.935; nilcsev sevem template2 contribution uk nilcsev = 1.125; nilcsev sevem coefficient error uk nilcsev = 0.7. The predicted-versus-published difference is the load-bearing entry, because it converts two separate error claims into one falsifiable joint statement.
Authors
- William Alexander Aldred (ORCID: https://orcid.org/0009-0000-1215-577X)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-16
- DOI
- https://doi.org/10.5281/zenodo.24270486
- Primary Topic
- Cosmology and Gravitation Theories
- Type
- article
- Field-Weighted Citation Impact
- 0.00