Autumn PDO gave no clear added winter precipitation contrast forecast gain in four western states in 1981-2025

We tested whether autumn PDO adds winter precipitation contrast forecast value after autumn ENSO is known. The target is the equal-state mean relative precipitation in Washington and Oregon minus that in California and Arizona. Each state uses a 1951-1980 winter reference. Fits use earlier winters only. For 45 winters in 1981-2025, adding PDO and its product with ENSO raised mean squared error by 2.71%. The conditional 95% five-year block interval for error reduction was -12.91% to +3.96%. Additive PDO, earlier predictors, recent training and a year term also left the gain uncertain. The result gives no clear added gain for the tested models and contrast. It does not show no physical PDO effect. Revised products, short dependent records, state averages and conditional intervals limit inference. Code, exact source hashes, aggregate evidence, an independent audit and a figure are included. External raw data and record tables stay outside the publication and can be regenerated.

PDO · ENSO · winter precipitation · western United States · chronological forecasts · null result