Twenty-degree averaging weakened the Pacific total steric variance reduction in 2005-2025 NOAA estimates

We tested how spatial averaging changes the total-to-thermal steric sea level variance ratio in NOAA upper-700-m fields. We used 84 seasons from 2005 to 2025 and removed linear trends and seasonal means. In the Pacific, the ratio rose from 0.860 at one degree to 0.923 at twenty degrees. The primary log ratio contrast was 0.071, with conditional 95% three-year block limits of 0.034 to 0.093. Thus, averaging weakened the variance reduction before the whole-basin scale. The twenty-degree ratio remained below one, with limits of 0.863 to 0.976. Wider regions did not give a simple monotonic change. The Atlantic ratio rose to 1.132 at sixty degrees but fell to 0.728 for its basin mean. Region boundaries and trend removal changed some secondary results. This extends a prior cell-to-basin test with intermediate regions. It uses the same source data and is not independent confirmation. One mapped product, twenty-one years and conditional uncertainty limit inference. The result does not identify a physical salinity feedback or coastal flood risk.

steric sea level · spatial averaging · variance · Pacific · salinity compensation · spatial scale