February western Pacific warm water volume gave no clear added winter ENSO forecast gain in 2000-2024

We tested whether February western Pacific warm water volume improves next-winter Nino 3.4 forecasts beyond current surface temperature, recent change, and an absolute centered-strength term. We used NOAA PMEL GODAS volume and NOAA CPC ERSSTv5 surface temperatures. Expanding fits used only earlier eligible annual origins. In 25 test years from 2000 to 2024, the western addition raised mean squared error from 1.0949 to 1.4188 C squared, or 29.6%. The paired gain was -0.3239 C squared, with a conditional 95% three-year block interval of -0.9499 to 0.2127. Basin volume gave a small uncertain gain. The western model had higher error than the basin model, with a conditional interval below zero for that secondary comparison. The primary interval includes both improvement and harm. Other issue months and baseline checks gave no clear western gain. These results test a small retrospective forecast model. They do not establish a physical cause or show that western heat content lacks value at other leads. Short records, final data snapshots, model form, and conditional uncertainty limit inference. The package includes exact source bytes and hashes, all forecasts, code, uncertainty draws, an independent audit, and a figure.

ENSO · warm water volume · western Pacific · forecast validation · GODAS · Nino 3.4 · uncertainty