Infinity Engine — Locked Pre-Registered Forecast for Solar Cycle 25 sunspot number (2026-04-11, version 1.1.0)

About this record: This is a benchmark-template publication from the Infinity Engine. Every Infinity record uses the same 12 sections in the same order with the same classification labels, so reviewers can navigate instantly. Every claim in this description is traceable to a file in the upload, with SHA-256 hashes provided in section 11. Classification labels used in this record: [MEASURED] direct observation, [METHODOLOGY] method itself, [INFERRED] deterministic derivation, [VALIDATED] instrument check passed, [PRE-REGISTERED] forward prediction awaiting falsification, [SYNTHESIS] interpretation across results, [HYPOTHESIZED] asserted but not yet validated. 1. Abstract [PRE-REGISTERED] This record contains a locked, SHA-sealed, falsifiable forward prediction for Solar Cycle 25 sunspot number, generated by the Infinity Engine's harmonic-regression forecaster on 2026-04-11. The forecast extends 12 months into the future from the end of the fit window, with bootstrap 95% prediction intervals and a SHA-256 lock hash that any reviewer can re-compute to verify the prediction has not been altered after publication. The prediction is checkable month-by-month against future data releases. The published numbers are the prediction; there is no escape clause. 2. Method Statement [METHODOLOGY] The forecaster is regularized harmonic regression with bootstrap prediction intervals. It is not a neural network, not a machine-learning black box, and uses no solar physics whatsoever. Every coefficient is traceable to a closed-form least-squares fit on the input time series. Locked configuration: fit_forecaster ncomp=3 poly=0 period_min=730 period_max=20000 clip_lower=0 Fit R²: 0.7151 (in-sample goodness of fit; not out-of-sample performance) Spectral components recovered: Period: 10.80 y, amplitude: 75.02 Period: 8.62 y, amplitude: 29.96 Period: 21.87 y, amplitude: 9.93 The configuration was selected after honest hindcast tests on past cycles to characterize the method's natural error envelope. See the side-by-side observation companion record for the full hindcast results. 3. Data Provenance [MEASURED] Input data: SILSO daily V2.0, monthly mean, ending 2026-03-15 The SILSO daily total sunspot number V2.0 is the canonical international reference for solar activity, maintained by the Royal Observatory of Belgium / SIDC under license CC-BY-NC 4.0 (Clette & Lefèvre 2016). It is fetched live by the Infinity engine via urllib + certifi from https://www.sidc.be/SILSO/DATA/SN_d_tot_V2.0.txt and cached locally after first fetch for reproducibility. The daily series is aggregated to monthly mean for this fit, which is the standard convention for Schwabe-cycle analysis (Hathaway 2015). 4. Headline Result [PRE-REGISTERED] The forecaster predicts the next solar minimum (cycle 25 → 26 boundary) at SN ≈ 0 around 2028-03-04. SHA-256 lock: 661365300298254985a47f776fd12a5243de7221664677fb697eaab7df2d328d Anyone can re-compute this hash from the lock file in this record and verify that the prediction has not been altered after the locked-at timestamp. 5. Full Results [PRE-REGISTERED] Every forecast point in the locked prediction: Months aheadForecast datePredicted SN95% CI lower95% CI upper32026-0698.65591.3048106.767262026-0982.642475.395490.75892026-1266.644759.632175.1343122027-0351.045743.894559.96152027-0636.221728.963245.2612182027-0922.531415.75331.7611212027-1210.30663.587819.5135242028-03008.8351272028-06000.6407302028-08000332028-11000362029-02000 6. Reproducibility [METHODOLOGY] Anyone with the source code can re-run this exact computation and verify the result matches the file in this record. Source file: infinity/forecaster.py Command: python -m infinity.forecaster # full self-test # Or programmatically: from infinity.real_data_sources import fetch_silso_sunspot from infinity.forecaster import fit_forecaster, extrapolate, clip_forecast_to_physical s = fetch_silso_sunspot(years=70) # SILSO daily, monthly aggregated below fc, c = fit_forecaster(s_monthly, n_components=3, poly_degree=0, period_min_days=730, period_max_days=20000) fc = extrapolate(s_monthly, fc, c, horizon_days=3*365.25, step_days=30, bootstrap_iters=300) fc = clip_forecast_to_physical(fc, lower=0) Random seed: 42 (fixed for full reproducibility) Dependencies: Python 3.12+ numpy certifi urllib (stdlib) The Infinity engine source code is available in the Infinity repository (linked in the related-identifiers section of this record). 7. Cross-References [METHODOLOGY] This record is a new version of an earlier locked forecast with concept DOI 10.5281/zenodo.19512977. The version lineage is preserved on Zenodo so any reader can find every refit under the same parent. This forecast record is part of a coherent four-record bundle characterizing the Infinity Engine's natural domain on cyclic phenomena: The 100-Test Validation Suite that established the engine's operating envelope on known signals The 16 Exploratory Experiments that mapped what the engine finds (and does not find) on open scientific questions The Side-by-Side Observation that shows what the same locked method produces on five documented forecasting events Companion records cross-linked on Zenodo (navigable in both directions): Haspart: 10.5281/zenodo.19513014 Haspart: 10.5281/zenodo.19513016 Haspart: 10.5281/zenodo.19513020 Version history: VersionDateDOIChange summary1.0.02026-04-1110.5281/zenodo.19512977Initial locked pre-registered forecast for solar cycle 25 declining phase and cycle 25 to 26 minimum. SHA-256: 661365300298254985a47f776fd12a5243de7221664677fb697eaab7df2d328d.1.1.02026-04-11(this record)Same locked numbers, same SHA seal. v1.1.0 adds hasPart cross-links to the three companion records (Validation Suite, Difficult Experiments, Side-by-Side Observation), adds version history table to section 7. No prediction values changed. 8. Falsifiability Protocol [PRE-REGISTERED] This is a forward prediction. It will be checked against future SILSO data releases as they arrive. The protocol: For each forecast point in the table above, the actual SILSO smoothed monthly sunspot number for that month, once published by the Royal Observatory of Belgium, will be compared to the predicted value. The prediction at a given lead time succeeds if the actual value falls within the combined 95% bootstrap interval (in the table) plus the ~30% method-error envelope characterized from hindcasts on cycles 23, 24, and 25. The cycle minimum prediction succeeds if a deep minimum (SN < 10) occurs within ±6 months of the predicted minimum date. No selective reporting. Every checkpoint will be reported as it arrives, success or failure. A new Zenodo record will be published periodically with the truth-vs-prediction comparison so the falsification trail is auditable. The SHA-256 lock above ensures the predicted numbers cannot be altered retroactively to improve the apparent fit. 9. Honest Limitations [METHODOLOGY] The reader should know exactly what this record does NOT claim: ~30% mean amplitude error on solar cycles, characterized from hindcasts. The 95% bootstrap CI is the fit uncertainty; the method error is wider. ~2 year mean timing error on the predicted minimum, characterized from hindcasts. No solar physics is used. No dynamo model, no magnetic precursor, no terminator method, no helioseismic input. Methods that use those (Dikpati flux-transport, McIntosh terminator, Schatten/Pesnell precursor) are in a different class and use information this record does not. Cycle 26 amplitude is not predicted. The method has no information about future cycle amplitudes beyond the average behavior of past cycles. The locked configuration was selected after seeing hindcast results, which introduces selection bias. Future predictions on other phenomena should lock the method before seeing test data. Negative-value clipping at 0 is applied because the unconstrained model produces unphysical negative sunspot numbers in the deep minimum. The model itself does not know SN ≥ 0 from physics. Any pitch or summary of this record that omits these limitations is misrepresenting the work. 10. Doctrine [METHODOLOGY] This record's framing is governed by the following durable doctrine principles, which apply to all Infinity publications regardless of kind: feedback_validation_vs_discovery.md — This is a forward prediction, not a discovery. Recovering known signals would be validation; predicting future ones is its own category that becomes science only when reality checks the answer. feedback_no_competition_framing.md — When this record is placed alongside published predictions from other groups, the framing is quiet observation only — no rankings, no winners, no scores. See the side-by-side observation companion record. 11. File Manifest [MEASURED] Every file in this upload, with size and SHA-256 hash. Any reviewer can download a file and verify it matches the hash here. If the hash differs, the file has been altered after publication and the description's claims may not apply. FilenameSizeSHA-256cycle25_LOCK_v2.json2.1 KB86dc65e03a0a79dc159824c38aaf86f36e63211f33e5bce3f4b192c4c03dbe96cycle25_locked_v2_latest.json3.9 KB44b1ae0fd425b1818b91110d3c240a6d1183fe0541c73a5f9308b7cb65db2127 12. Citation [METHODOLOGY] Prose citation: Sharma, Ishva Shailesh B. (2026). Infinity Engine — Locked Pre-Registered Forecast for Solar Cycle 25 sunspot number (2026-04-11, version 1.1.0). Zenodo. [DOI to be assigned at publish time] BibTeX: @dataset{infinity_2026, author = {Sharma, Ishva Shailesh B}, title = {Infinity Engine — Locked Pre-Registered Forecast for Solar Cycle 25 sunspot number (2026-04-11, version 1.1.0)}, year = {2026}, publisher = {Zenodo}, }