Hysteresis is a concept borrowed from physics and engineering, describing systems whose output depends not only on their current conditions but on the path they took to get there. In the climate system, it means that the world you get during a period of carbon removal may look very different from the world you had at the same CO2 level on the way up. For temperature, this asymmetry has been studied extensively in idealized model experiments. What has remained far less clear is how precipitation extremes, which depend on intricate interactions between moisture, circulation, and sea surface temperature patterns, behave when humanity eventually begins extracting carbon dioxide from the atmosphere at scale.

The research team tackled this problem using climate model simulations built around four different CO2 concentration thresholds and two distinct rates of emission reduction. In these pathway experiments, carbon dioxide concentrations first rise to a peak, then decline as emissions fall and removal proceeds, before stabilizing back at the initial concentration. By comparing the state of the climate during the ramp-up phase with the state during and after ramp-down at identical CO2 levels, the researchers could isolate the irreversibility signal from the direct response to greenhouse gas forcing. The design allowed them to test not just whether hysteresis occurs, but how its magnitude depends on the choices societies make about the timing and pace of decarbonization.

The results are striking. Extreme precipitation indices based on both intensity and frequency increase as CO2 climbs to its peak, which is exactly what physical understanding would predict, since a warmer atmosphere holds more water vapor and energizes storms through the Clausius-Clapeyron relationship. The surprise comes on the way down. When CO2 concentrations return to their initial values, the extreme precipitation indices do not. They remain elevated above their starting point, a residual enhancement that persists even though the greenhouse gas forcing that produced it has been nominally erased. In other words, the most dangerous aspect of East Asian climate change, the flooding rains of an intensified monsoon, would linger after the cause had been addressed.

Crucially, the study shows that this hysteresis is not a fixed property of the climate system but a variable one, amplified by how slowly humanity acts. When emission reductions are delayed or proceed at a slower rate, the lag between CO2 decline and precipitation recovery grows, and the irreversibility becomes more pronounced. This finding carries a direct policy implication: the sequencing of decarbonization matters as much as its endpoint. Two pathways that end at the same atmospheric CO2 concentration can leave East Asia with materially different risks of extreme rainfall depending on whether the peak was reached early and reduced quickly, or late and grudgingly. Carbon removal, in this sense, is not a clean slate but a partial one.

To understand why extreme precipitation gets stuck in its elevated state, the researchers traced the signal back to the tropical Pacific Ocean. The hysteresis in East Asian rainfall is linked to persistent changes in the pattern of tropical Pacific warming, which do not reverse in step with CO2 concentrations. Sea surface temperature patterns are notoriously sluggish, shaped by ocean heat uptake and circulation adjustments that operate on timescales far longer than the atmospheric response to greenhouse gases. Once the tropical Pacific settles into an altered warming configuration, that configuration continues to influence the atmosphere above it, generating teleconnections, vast waves of pressure and circulation anomalies, that propagate into the East Asian monsoon region and sustain enhanced mean monsoon precipitation long after the forcing has subsided.

The team then dissected the mechanism further using a moisture budget analysis, a technique that decomposes changes in precipitation into its thermodynamic and dynamic contributors. The thermodynamic factor concerns how much water vapor the air carries: warmer air holds more moisture, and in the early summer months the study finds that rises in specific humidity enhance the transport of moisture into East Asia through the mean atmospheric circulation. The dynamic factor concerns how the air moves: in late summer, the monsoon circulation itself intensifies, strengthening the winds that funnel moisture-laden air from the adjacent oceans onto the continent. Both effects contribute to the enhanced precipitation, and both are sustained by the hysteresis in the underlying ocean and circulation state, which explains why the extremes they feed remain elevated during the removal phase.

The regional stakes could hardly be higher. East Asia is home to some of the world’s densest populations and most productive agricultural systems, all calibrated to the rhythm of the monsoon. Extreme precipitation events in the region have repeatedly caused catastrophic flooding, landslides, and economic losses, and the study’s findings suggest that these risks would not simply unwind in parallel with a global decarbonization effort. Infrastructure designed, reservoirs managed, and flood defenses built around the assumption that climate risk scales linearly with CO2 would systematically underestimate the danger during the removal era. Adaptation planning, the research implies, must account for the possibility that the climate of the cleanup decades will be more hazardous than the concentration numbers alone would suggest.

More broadly, the work adds East Asian extreme precipitation to a growing list of climate impacts, from sea level rise to ice sheet loss, whose response to carbon removal is asymmetric and path-dependent. It underscores a message that has been building across the reversibility literature for years: preventing climate change is far more reliable than reversing it. The study’s comparison of emission reduction rates shows that every year of delay and every increment of slower action deepens the hysteresis and extends the period during which East Asia endures amplified extreme rainfall. The timing and pace of emission reductions, the authors conclude, play a critical role in determining regional climate outcomes, not merely the final concentration target.

As nations weigh the role of carbon dioxide removal in future climate portfolios, this research offers a sobering technical caveat. Negative emissions, whether through afforestation, direct air capture, or enhanced weathering, are increasingly treated as a safety net that can compensate for overshooting temperature goals. But the East Asian monsoon system, coupled as it is to the slow memory of the tropical Pacific, does not reset on demand. The downpours that define the peak-CO2 world cast a shadow across the removal decades, and only faster, earlier emission cuts can shrink that shadow. The atmosphere’s carbon ledger may be reversible on paper; the monsoon, this study shows, keeps its own account.

Subject of Research: Hysteresis of extreme precipitation responses to carbon dioxide removal pathways in East Asia

Article Title: Hysteresis of extreme precipitation in East Asia under various CO2 removal pathways

Article References: Hysteresis of extreme precipitation in East Asia under various CO2 removal pathways. (n.d.). https://doi.org/10.1038/s41467-026-78431-x

Image Credits: AI Generated

DOI: 10.1038/s41467-026-78431-x

Keywords: extreme precipitation, East Asia, CO2 removal, hysteresis, monsoon, tropical Pacific, climate modeling, moisture budget, emission reductions, climate reversibility, Nature Communications, atmospheric teleconnections