SF Winter 2026 · process · Friday 2 October 2026
NOAA ENSO Normals 2020 — Conditional Climatology, Not a New SF Forecast
Documentary review only. No local grid values imported; no model, score, October input, or issued prediction changed. Protocol first (c15e7e7); results 4c50e97.
Bottom line
- Product: historical monthly gridded conditional means and quantiles with secular-change adjustment — static, not operational.
- Climate data 1951–2020, recentered on 2006–2020 OCN; 1/24° CONUS land grid; TMAX/TMIN (°C) and liquid-equivalent precip (mm).
- Categories from traditional ONI tercile/sextile ranks with season-specific thresholds and persistence rules — not today’s RONI strength bands. Prototype DJF “strong” boundary ≈ +1.03°C is method-dependent, not a universal +1.5/+2.0 definition.
- Monthly quantiles cannot simply be added into a winter interval; rain anomalies live in percent-of-normal space; TMAX/TMIN quantiles ≠ TAVG quantile without a joint law.
- Useful context: represent secular change and broad conditional variability; do not mistake a mean composite for a guaranteed outcome.
What this does not do
Does not justify a station rainfall multiplier, identified ENSO causal increment, replacement temperature trend, calibrated DJF interval, daily storm/AR count, or any revision of the September 30 v1 forecast. Release vs 2019 prototype OCN windows (paper 10-yr temp / 15-yr precip endpoints vs release 15-yr 2006–2020) must not be silently collapsed. No NetCDF downloaded; no local anomaly computed.
Note: enso-normals-documentary-review.md. Prior SF desks: CPC verification tip 7375; daily temp calendar/transport audit tip 7385. v1 immutable.