The Nino 3 minus Nino 4 contrast gave no clear added winter PDO forecast gain in 1981-2025

We tested whether the summer Nino 3 minus Nino 4 temperature anomaly adds information about winter PDO after recent PDO and Nino 3.4 are known. Fixed NOAA ERSSTv5 tables supplied 74 annual rows. Each forecast used earlier years only. Across 45 test years in 1981-2025, adding the contrast raised winter mean squared error from 0.572125 to 0.591562, an increase of 3.40%. The conditional 95% five-year block interval for error reduction was -15.15% to +5.31%. Summer error rose by 6.93%. Winter checks with 30-year fits, a time trend and lagged Nino 3.4 also had negative point skill and intervals across zero. This test gives no clear added linear forecast benefit. It does not show that tropical heating location has no physical effect. Short dependent records, revised surface indices and fixed-prediction uncertainty limit the result. The package includes exact inputs, hashes, all predictions, code, a numerical audit and a figure.

PDO · ENSO · tropical Pacific · spatial contrast · seasonal prediction · null result · chronological validation