The Madden–Julian Oscillation (MJO) has important socioeconomic impacts due to its influence on both tropical and extratropical weather extremes. In this study, we use machine learning (ML) to correct the predictions of the weather model holding the best performance, developed by the European Centre for Medium-Range Weather Forecasts (ECMWF). We show that the ML post-processing leads to an improved prediction of the MJO geographical location and intensity.

The dynamics of our climate involves multiple timescales, and while a lot of work has been devoted to quantifying variations in time-averaged variables or variations in their seasonal cycles, variations in daily variability that occur over several decades still remain poorly understood. Here we analyse daily surface air temperature and demonstrate that inter-decadal changes can be precisely identified and quantified with the Hilbert analysis tool.

We analyze the dynamical and statistical properties of two surface air temperature (SAT) reanalysis datasets. For each SAT time-series we analyze i) the distance between the lagged SAT time series and the insolation, and ii) the Shannon entropy computed from the probability distribution function (pdf) of SAT values. We show that these simple measures uncover meaningful long-range coherent spatial structures that emerge from the local properties of SAT time-series.

Understanding how surface air temperature (SAT) is controlled by the incoming solar radiation is a fundamental and challenging problem in climate dynamics. Here we analyze the response of monthly-averaged SAT to solar forcing, and we also quantify the level of randomness of SAT variability. We find coherent spatial patterns which can be interpreted in terms of known climate phenomena.