Development and Application of Two New Atmospheric Indices for Monitoring and Surveillance of the El Niño–Southern Oscillation in Colombia

ABSTRACT

Two atmospheric indices are introduced to improve the monitoring and early detection of El Niño–Southern Oscillation (ENSO) variability over Colombia: the Inter‐Basin Pressure Gradient Index (IPGI) and the Zonal Moisture Index (ZOMI). IPGI is constructed from monthly mean sea‐level pressure anomalies between the eastern tropical Pacific and the Caribbean Sea, while ZOMI is based on anomalies of vertically integrated horizontal moisture flux over Colombia. In both cases, the underlying fields are linearly detrended through the removal of the least‐squares linear trend, standardised, and low‐pass filtered with a second‐order Butterworth filter using a 1/12 cutoff frequency to isolate interannual variability. Using monthly data for the period 1981–2024, the indices are evaluated through cross‐correlation analysis, canonical correlation analysis (CCA), and Monte Carlo AR(1) significance testing to assess their statistical robustness and consistency with large‐scale ENSO variability. The results indicate that IPGI and ZOMI capture distinct but complementary modes of coupled ocean–atmosphere variability associated with the diversity of ENSO regimes across the equatorial Pacific. Cross‐correlation analyses show that the atmospheric indices systematically lead commonly used ENSO sea surface temperature (SST) indices by approximately 1–3 months, highlighting their potential as early atmospheric indicators. Autocorrelation‐aware significance testing reveals a scale‐dependent coupling structure: in the non‐prewhitened analysis the second canonical mode shows robust significance and reflects persistent low‐frequency coupled variability, whereas after prewhitening the leading ENSO‐related mode becomes statistically significant, indicating a more immediate SST–atmosphere interaction. Compared with traditional ENSO indices (ONI, MEI, SOI, Niño regions), IPGI and ZOMI provide earlier and more regionally relevant representations of ENSO‐related atmospheric variability, underscoring their value for climate monitoring, surveillance, and early warning applications over Colombia.