ERSST

ERSST, or Extended Reconstruction Sea Surface Temperature, is a global gridded dataset that provides historical reconstructions of sea surface temperature (SST) based on in-situ observations. It is widely used in climate research and monitoring to extend SST records back before the satellite era and to support analyses of long-term temperature trends and variability.

The dataset is produced from the International Comprehensive Ocean–Atmosphere Data Set (ICOADS) observations, including ships, buoys, and other platforms. To create a consistent long-term record, ERSST applies quality control and bias adjustments to account for inhomogeneities introduced by changes in measurement methods (such as bucket versus engine intake), vessel type, and observation practices. After corrections, the observations are merged and interpolated to regular grids (commonly near 2°x2°) to produce global gridded SST fields. Anomalies are typically computed relative to a baseline period (for example 1901–2000), enabling comparison across time.

Several versions of ERSST exist, with v3b, v4, and v5 representing successive improvements in data coverage, bias-correction procedures, and methodological refinements. Each update aims to reduce non-climatic variability and to align more closely with other SST records and the satellite era.

Applications of ERSST include climate trend analyses, global mean surface temperature calculations, and studies of regional phenomena such as El Niño–Southern Oscillation and the Atlantic Multidecadal Oscillation. The dataset is maintained and distributed by NOAA, and is a standard reference in climate literature and IPCC assessments. Limitations include sparser coverage in the 19th and early 20th centuries and residual biases in early records.