Hidden Pacific Pattern Extends Summer Monsoon Forecasts to Four Years

Every summer, the East Asian monsoon delivers the rainfall that sustains hundreds of millions of people across China, Korea, Japan, and the surrounding region. When the monsoon rainband shifts or weakens, the consequences ripple through agriculture, water reservoirs, flood defenses, and energy planning. Yet despite decades of research, seasonal forecasts of the East Asian summer monsoon remain frustratingly limited, and predictions beyond a single season have long been considered near-impossible. A new study published in Climate Dynamics now suggests that a previously underappreciated source of multi-year predictability has been hiding in plain sight in the Pacific Ocean, one that operates independently of the famous El Niño and Southern Oscillation cycle.

The research, conducted by Hak-Jun Lee of GeoSystem Research Corporation, Sang-Wook Yeh of Ewha Womans University, and Myong-In Lee of Ulsan National Institute of Science and Technology, focuses on a Pacific sea surface temperature pattern known as a tripole. This pattern, familiar to climate scientists from studies of the Interdecadal Pacific Oscillation, consists of anomalies in three key basins: the Northwest Pacific, the central equatorial Pacific, and the Southwest Pacific. When one region is warmer than average while the other two are cooler, and vice versa, the ocean surface takes on a three-poled structure that can persist for years and influence atmospheric circulation far beyond the tropics.

The challenge the researchers faced is that this tripole pattern does not exist in isolation. El Niño and its cold counterpart La Niña, together comprising the El Niño and Southern Oscillation, or ENSO, imprint their own signature on Pacific sea surface temperatures and dominate the tripole signal. To isolate the part of the variability that operates independently, the team constructed what they call a linear ENSO-independent tripole index, abbreviated EITI. The method is statistically elegant: they computed the tripole index from sea surface temperature anomalies averaged over the three Pacific regions during boreal summer, then subtracted the portion of that index that could be linearly explained by the simultaneous Niño3.4 index, the standard yardstick of ENSO strength measured in the central equatorial Pacific. What remains is a clean measure of Pacific variability that has nothing to do with concurrent ENSO conditions.

Using observed sea surface temperatures from the NOAA Extended Reconstructed Sea Surface Temperature version 5 dataset, together with precipitation data from the Global Precipitation Climatology Project and atmospheric reanalysis from the Copernicus ERA5 product, the researchers examined what happens during the positive phase of the EITI, when the tripole pattern reaches a particular configuration. The results were striking. A positive EITI phase is associated with enhanced precipitation along the East Asian monsoon rainband, meaning wetter conditions across the densely populated monsoon region. The atmospheric machinery behind this response involves two well-known components of the summer circulation: the subtropical jet stream and the western North Pacific subtropical high.

During positive EITI phases, the team found that the subtropical jet shifts equatorward, toward the equator, while the western North Pacific subtropical high strengthens. Both changes matter enormously for the monsoon. The subtropical jet acts as a guide rail for storm tracks and is intimately tied to the position of the meiyu-baiu rainband that stretches from eastern China through Korea and Japan; a shift in the jet translates directly into a shift in where the heaviest summer rains fall. The subtropical high, meanwhile, is the great anticyclonic circulation over the western Pacific that steers moist air masses toward East Asia and controls the timing and intensity of the monsoon onset. When the high strengthens, it pumps more moisture-laden air toward the rainband, amplifying rainfall. The EITI thus provides a single oceanic index that captures the coordinated behavior of these two circulation systems.

But the most consequential finding of the study concerns predictability rather than mechanism. The researchers turned to the Decadal Climate Prediction Project component of the sixth Coupled Model Intercomparison Project, known as CMIP6 DCPP. This international effort coordinates hindcast experiments in which state-of-the-art climate models are initialized with observed ocean and atmosphere conditions and then run forward for several years, allowing scientists to test how far ahead different climate features can actually be predicted. The team evaluated how well the models predicted the winter Niño3.4 index and the summer EITI at various lead times, comparing the forecasts against observations.

The comparison revealed a sharp contrast between the two indices. The boreal winter Niño3.4 index, the classic ENSO measure, is skillfully predicted at a lead time of one year, but its prediction skill declines rapidly as the lead time increases. This mirrors the well-documented spring predictability barrier and the fundamental limits of ENSO forecasting, which rarely extends useful skill beyond a year. The summer EITI, by contrast, remains skillfully predicted up to a lead time of four years. Because the EITI is independent of ENSO, its persistence reflects the slower, longer-lived dynamics of the broader Pacific climate system, including the decadal-scale ocean memory that underlies patterns such as the Pacific Decadal Oscillation and the Interdecadal Pacific Oscillation.

Even more importantly, the link between the EITI and the East Asian summer monsoon survives the forecasting test. The researchers found that the EITI and EASM relationship is most robustly reproduced in the multi-model ensemble, the combined output of many different climate models, and that this relationship remains statistically significant at lead times of up to three years. In other words, when the models collectively predict the state of the ENSO-independent Pacific tripole several years ahead, that prediction carries genuine information about the likely behavior of the East Asian summer monsoon. This is a remarkable result for a climate feature that has historically resisted prediction beyond a single season, and it suggests that the multi-model ensemble approach, which averages out the idiosyncratic errors of individual models, is particularly effective at capturing this slow ocean-atmosphere coupling.

The implications extend well beyond the academic literature. Multi-year lead forecasts of monsoon rainfall would transform water resource management across East Asia, allowing reservoir operators, agricultural planners, and disaster preparedness agencies to anticipate prolonged wet or dry phases years in advance rather than reacting to seasonal forecasts issued only months ahead. The study also adds to a growing body of work showing that components of the climate system other than ENSO, including the Indian Ocean basin mode, the tropical Atlantic, and Pacific decadal variability, contribute substantially to East Asian climate. By explicitly removing the ENSO contribution, the new EITI framework clarifies how much of the monsoon’s variability is governed by these slower, more predictable oceanic patterns.

Caveats remain, as they always do in climate science. The analysis relies on linear regression to strip out the ENSO signal, and nonlinear interactions between ENSO and the tripole pattern could complicate the picture in ways a linear framework does not capture. The robustness of the EITI and EASM relationship also varies among individual models, appearing most reliably only in the multi-model ensemble, a reminder that model biases in simulating Pacific sea surface temperatures and monsoon dynamics still limit how confidently these results can be translated into operational forecasting. Nevertheless, the core message of the study is clear and potentially transformative: the Pacific Ocean stores climate information that persists for years, and by learning to read the ENSO-independent part of that signal, scientists may finally extend the horizon of East Asian summer monsoon prediction from months to years, offering societies across the monsoon domain a far longer window of preparation for the rains that shape their lives.

Subject of Research: ENSO-independent Pacific sea surface temperature variability and its multi-year predictability for the East Asian summer monsoon

Article Title: The influence of linear ENSO-independent Pacific variability on the East Asian summer monsoon and its predictability

Article References: Lee, H.-J., Yeh, S.-W., & Lee, M.-I. (2026). The influence of linear ENSO-independent Pacific variability on the East Asian summer monsoon and its predictability. Climate Dynamics, 64(11), Article 445. https://doi.org/10.1007/s00382-026-08398-4

Image Credits: AI Generated

DOI: 10.1007/s00382-026-08398-4

Keywords: East Asian summer monsoon, ENSO, Pacific tripole, sea surface temperature, CMIP6, decadal climate prediction, subtropical jet, western North Pacific subtropical high, multi-year predictability, Climate Dynamics, Pacific Decadal Oscillation, climate modeling