New direct measurements from anchored ocean buoys provide the strongest observational evidence yet that the Atlantic Meridional Overturning Circulation (AMOC) is weakening across multiple latitudes in the western Atlantic. The peer-reviewed study led by Qianjiang Xing at the University of Miami, reported in New Scientist, analyzed data from 2004 to 2023 collected by the RAPID-MOCHA array stretching from the Bahamas to the Canary Islands, supplemented by three additional western-boundary mooring arrays off the West Indies, the US East Coast, and Nova Scotia. These arrays continuously record temperature, salinity, and velocity; researchers convert pressure differences into estimates of water transport. The data show a decline of roughly 90,000 cubic meters per second per year, equating to about 10 percent weakening over two decades, with tighter uncertainty bounds at the western sites than the full-basin array.

This goes beyond earlier proxy-based estimates. A landmark 2018 Nature paper by Caesar et al. used sea-surface temperature 'fingerprints' to infer a roughly 15 percent slowdown since 1950; the new work supplies the missing direct velocity confirmation at depth. It also aligns with a 2024 Science Advances study by van Westen and colleagues, which employed high-resolution climate models to identify an early-warning signal suggesting the AMOC may be approaching a critical freshwater threshold that could trigger collapse within decades under high-emission scenarios. The current observational record (19 years of continuous in-situ data at four latitudes) is still short relative to natural decadal variability driven by phenomena such as the North Atlantic Oscillation, a limitation the authors explicitly note. Nonetheless, the spatial coherence across independent arrays markedly reduces the chance that the trend is merely an artifact of local noise.

Mainstream coverage, including the New Scientist piece, correctly highlights Greenland meltwater as the likely culprit: freshwater dilutes surface density, slowing the sinking that powers the southward return flow. What it underplays is the web of connected tipping risks. Paleoclimate records show that past AMOC disruptions, such as during the Younger Dryas, triggered abrupt Northern Hemisphere cooling of 5–10 °C within decades while shifting tropical rainfall belts. Today, a weakening AMOC would raise sea levels an extra 0.5–1 meter along the US East Coast and Maritime Canada by reducing the dynamic pull on ocean water, independent of overall steric rise. It would also warm the Southern Ocean, accelerating Antarctic ice-shelf melt and potentially destabilizing the West Antarctic Ice Sheet. Monsoon systems in Asia and West Africa could shift, threatening food production for billions.

The original reporting also gives insufficient weight to compounding feedbacks. Rapid Arctic sea-ice loss and increased Greenland runoff are already documented; an AMOC slowdown would further warm high-latitude seas, releasing more methane from permafrost and seabed hydrates. Marine ecosystems are already showing stress: the Gulf Stream's north wall is shifting, altering fish migration routes and hurricane intensity. Computer models have long projected AMOC decline, yet only in the past few years have observational arrays reached the length needed to detect a robust anthropogenic signal amid natural variability.

The study methodology is state-of-the-art for physical oceanography but not infallible. Mooring arrays provide high temporal resolution at fixed points; extrapolating to full-basin transport still involves assumptions about geostrophic balance. Sample size is necessarily limited by the high cost of sustained deep-ocean instrumentation; two decades cannot fully rule out multi-decadal oscillations. These caveats matter. Yet the convergence of proxy studies, direct measurements, and process-based models paints a consistent picture that mainstream climate discourse has treated too cautiously.

Collectively these sources indicate we are not merely observing gradual change but approaching a potential bifurcation point. Continued emissions push us closer to the threshold where recovery becomes impossible on human timescales. Expanded monitoring, including new arrays at additional latitudes and better satellite gravimetry, is essential. Policymakers and the public need far deeper analysis of these cascading risks rather than repeated reassurances that catastrophe remains centuries away. The buoys are sounding an alarm that deserves far wider attention.