The El Nino Southern Oscillation (ENSO) is the most robust, coupled ocean-atmospheric component of intraseasonal-interannual variability on the globe. Therefore, an index that effectively characterizes a large fraction of ENSO’s total variability is deemed to be of upmost importance. This work uses sea surface temperature (SST) and sea level pressure (SLP) datasets to produce a major update to the original Extended Multivariate ENSO Index (MEI.ext; Wolter and Timlin, 2011) that we call the Extended Multivariate ENSO Index version 2 (MEI.extv2). MEI.extv2 covers a timespan of 1865-2020 at a monthly time resolution where each month represents a rolling bi-monthly average, and we include accompanying estimates of the confidence intervals. By utilizing an ensemble of quality-controlled reanalysis datasets, creating a statistical suite of MEI "realizations", and applying a weighting procedure that accounts for time-varying quality and quantity of the underlying SST and SLP data since the mid-1800s, we argue that the MEI.extv2 is more robust, stable, and coherent than the original MEI.ext. A novel contribution of this work is to provide an empirical estimate of the uncertainty associated with the MEI that is built from the time-varying uncertainties apparent across the input reanalysis datasets. Overall, MEI.extv2 agrees well with the original MEI.ext, showing that ENSO was more active near the turn of the 20th and 21st centuries, but there are differences in MEI.extv2 and MEI.ext for individual ENSO events. We use our uncertainty estimates to assess conclusions about the duration, amplitude, spacing, and the onset and decay of ENSO on bi-monthly temporal scales. MEI.extv2 provides a comprehensive ENSO index that extends from 1865 to the present day, includes a novel estimate of uncertainty, and is tied to observationally based records that themselves continue to be updated.
