Teleconnections act like invisible threads linking tropical anomalies to weather in the mid-latitudes. Related papers in our archive address the predictability of El Niño, the ocean conveyor belt, and solar cycles, but no one has systematically tested whether the threads themselves are changing under greenhouse gas forcing. This study fills that gap by applying formal attribution to covariance matrices of temperatures.
The authors used two approaches. The first, covariance fingerprinting, projects the observed change in the covariance matrix of monthly temperatures onto the change simulated by an ensemble of models with historical forcing. The second, pointwise multiple linear regression, estimates the contribution of each of five climate modes (ENSO, NAO, SAM, IOD, PDO) to temperatures for two 30-year periods. Significance was assessed by comparison with 1000 random pairs of periods from a control run of the CESM2 model.
The observed shift in covariance structure between 1960–1990 and 1990–2020 was larger than any of the 1000 random shifts in the control run. Pointwise analysis showed significant teleconnection changes for ENSO on 20% of grid cells, for IOD on 15%, and for GMST on 28%. The variance of the modes themselves barely changed, pointing to a change in the wave propagation environment rather than in the intensity of the sources.
Unlike related studies that examined predictability or individual regions, this work is the first to formally attribute a global shift in teleconnections. This changes the perspective on adaptation: regional climate risks depend not only on mean warming but also on the reorganization of connections. Other studies in the archive address noise and signal in climate but do not provide quantitative attribution of such reorganizations.
Future studies will likely separate the contributions of altered circulation and altered humidity to the strengthening of teleconnections. More realistic emission scenarios than RCP 8.5 may be used. It is also promising to study the impact of reorganized teleconnections on extreme events, not just mean temperatures.
The results will affect regional climate planning, seasonal forecasting, and risk assessment for agriculture and water resources.
The next step could be a mechanistic explanation of the changes, for example through analysis of Rossby wave propagation in an altered mean flow.
This work directly relates to the unresolved problem of separating internal variability from forced response in the observed climate. It also touches on the question of how stable statistical relationships used in empirical forecasting are.
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
The study shows that shifts in remote atmospheric connections (teleconnections) between 1960–1990 and 1990–2020 are not explained by natural variability. The authors applied covariance fingerprinting and pointwise regression to observational and model data, establishing that anthropogenic influence has already reshaped the impacts of El Niño, the Indian Ocean Dipole, and the Pacific Decadal Oscillation on regional temperatures.
Where its strength is
The strength of the work lies in the formal attribution of the global shift in teleconnections using two independent methods and rigorous significance testing against a control run. This allows separating the anthropogenic signal from natural variability and provides a quantitative basis for assessing changes in regional climate risks.
Where it could go next
Worth trying
Why this may matter
Understanding how anthropogenic influence reshapes teleconnections is critical for adaptation: regional climate risks can change not only due to mean warming but also due to altered connections between distant regions. This could affect forecasting of droughts, floods, and other extreme events.
Nearby works in our archive
The archive is sparse around this work: the closest review is only 0.51, against a usual 0.64. It looks like few have drilled nearby.
The sign change of the ENSO teleconnection in northeastern Africa means that El Niño, which used to bring warmth there, may now bring cooling.