El Niño is the strongest natural climate oscillation on interannual timescales, affecting weather worldwide. It is described by various indices, but MEI is considered the most comprehensive because it combines six meteorological parameters over the tropical Pacific. Related studies have shown that climate subsystems synchronize as a network of coupled nonlinear oscillators: for example, ocean temperature, day length, and auroras share a common 60-year cycle. This study stands somewhat apart: instead of looking for cycles, it directly measures the strength of coupling between three indicators.
Monthly data for three indices were used: MEI (1950–2011), CO₂ concentration at Mauna Loa (1958–2011), and day length LOD (1962–2010). First, seasonal cycles were removed from CO₂ and LOD by computing year-over-year differences (difference between January of the current year and January of the previous year, etc.). Then, cross-correlations between MEI and the resulting annual rates of change of CO₂ and LOD were calculated, varying the time shift. To quantify the relationship, scatter plots were constructed and linear conversion coefficients were determined.
A significant correlation was found between MEI and the annual rate of change of CO₂ (r = 0.49 with a 3-month lag for CO₂) and between MEI and the annual rate of change of LOD (r = -0.34 with a 4-month lag for LOD). Both correlations are statistically significant (p < 0.01). This means that after a strong El Niño (high MEI), the CO₂ growth rate peaks 3 months later, and the rate of change of LOD reaches a minimum 4 months later (i.e., Earth's rotation accelerates). An exception is the period 1991–1993, when the Pinatubo eruption disrupted the synchrony due to reduced ecosystem respiration and increased diffuse radiation. The conversion coefficients are: annual CO₂ rate ≈ 0.55 × MEI (ppm/year), annual LOD rate ≈ -0.27 × MEI (ms/year).
This work differs from related studies in that, instead of searching for common cycles, it provides simple quantitative relationships that can be directly used to test models. If a model reproduces these coefficients, it correctly describes the coupling between the ocean, atmosphere, and Earth's rotation. This is a first step toward turning empirical correlations into a validation tool.
The topic could evolve toward incorporating astronomical influences (solar and lunar tidal cycles) into El Niño models. It is also possible to extend the analysis to other climate indices, such as the North Atlantic Oscillation, to identify a broader synchronization network. Further refinement of the coefficients on longer data records would allow testing their stability over time.
The results could influence climate modeling, geophysics of Earth's rotation, and the carbon cycle by providing simple tests for models.
The next step could be to test the obtained coefficients on independent data and within existing climate models to assess their predictive power.
This work relates to an unsolved problem in climatology: why general circulation models poorly reproduce El Niño and associated oscillations. The proposed quantitative relationships could serve as a benchmark for improving models, but they do not resolve the question of the physical mechanism generating El Niño.
The eruption of Mount Pinatubo in 1991 so strongly affected the global carbon cycle that it disrupted the link between El Niño and the CO₂ growth rate for two years.