Understanding today's climate requires explaining two phenomena: the trend in global temperature and the El Niño–Southern Oscillation (ENSO), a large-scale redistribution of warm water in the tropical Pacific. Climate models diverge from observations: since 1998, the rise in annual mean temperature has been statistically insignificant, and ENSO forecasts are inconsistent. This work aligns with research linking climate variability to the biosphere and the carbon cycle, but differs by considering not only the rate of change of CO₂ but also its acceleration.
The authors used monthly data from 1960 to 2012: CO₂ concentration from Mauna Loa, tropical temperature (30°N–30°S) from the HadCRUT4 dataset, the Southern Oscillation Index (SOI), volcanic aerosols, and solar radiation. All series were normalized to zero mean and unit variance and smoothed with a 13-month moving average. The first and second derivatives of CO₂ were computed via finite differences. Correlation analysis with time lags and multiple linear regression were used to assess relationships.
The first derivative of CO₂ is in phase with tropical temperature, while the SOI leads both by two months. The second derivative of CO₂, in contrast, leads the SOI and temperature. The correlation between the second derivative of CO₂ and the SOI was 0.315, higher than that of the SOI with any other factor. In the regression model, the first derivative of CO₂ explains temperature as well as the standard set of four predictors (CO₂ level, SOI, volcanoes, sun), while the volcanic factor becomes insignificant. After correcting for autocorrelation, the relationships remain statistically significant: p = 0.0013 for the pair “first derivative of CO₂–temperature” and p = 0.001 for the pair “second derivative of CO₂–SOI”.
This work follows the general line of neighboring studies but adds a new element: it considers the acceleration of CO₂ change, not just its level or rate. This allows ENSO to be explained as a response of the biosphere to information about the rate of change of CO₂, rather than as an independent ocean mode. This view ties together the carbon cycle, vegetation, and large-scale atmospheric circulation.
The topic could develop toward testing the hypothesis that plants are sensitive to the acceleration of CO₂ change. Experiments on plant physiology would be needed to show whether they can detect the second derivative. It is also important to test the robustness of the correlations over longer records and with other climatic factors accounted for.
The results could influence climate models by incorporating a biospheric response to the rate of change of CO₂. This also pertains to El Niño forecasting and understanding the causes of the warming pause.
The next step is to reproduce the analysis on independent data and check whether the lead of the second derivative of CO₂ over the SOI holds in other periods. It is also worth investigating possible mechanisms by which plants perceive the acceleration of CO₂ change.
This work touches on a fundamental problem in climatology: why models do not reproduce the observed warming pause. It proposes a new predictor, the rate of change of CO₂, which could improve models. Furthermore, it links ENSO to the biosphere, opening a path to solving the long-standing puzzle of the El Niño mechanism.
The energy that the biosphere obtains through photosynthesis is about 130 terawatts, which is 10–20% of the power of all winds in the atmosphere. That means plants can influence climate not only chemically but also through energy and water fluxes.