El Niño-Southern Oscillation (ENSO), the strongest air-sea coupling system in the tropics, significantly impacts global climate, ecosystems, and human societies. Understanding ENSO is crucial for improving climate prediction. While ENSO's tropical dynamics are well-studied, recent research highlights the influence of extratropical variability. The boreal winter North Pacific Oscillation (NPO), a major atmospheric mode over the North Pacific, is a key player, impacting ENSO development through a seasonal footprinting mechanism. Positive winter NPO leads to SST warming in the subtropical North Pacific, which propagates southwestward, inducing westerly wind anomalies in the equatorial western-central Pacific via the wind-evaporation-SST (WES) feedback. This ultimately influences ENSO evolution. Observations show a strong correlation between winter NPO events and subsequent ENSO events, including major El Niño and La Niña events. The enhanced impact of winter NPO since the early 1990s has been linked to the increased frequency of central Pacific El Niños and multi-year El Niño events. However, the impact of global warming on this NPO-ENSO relationship remains unclear. This study addresses this gap by using climate model simulations to investigate how the influence of winter NPO on ENSO changes under greenhouse warming, contributing to a better understanding and prediction of future ENSO events.

Several studies have explored the extratropical influences on ENSO, particularly focusing on the North Pacific Oscillation (NPO) and its seasonal footprinting mechanism. Vimont et al. (2001, 2003) demonstrated the connection between mid-latitude atmospheric variability and tropical ENSO development. Other studies (Yu & Kim, 2011; Chang et al., 2007; Vimont et al., 2014) investigated the relationship between extratropical patterns like the Pacific Meridional Mode (PMM) and ENSO characteristics, such as the central-Pacific and eastern-Pacific types of ENSO. The role of the NPO in modulating ENSO's complexity and the increasing prevalence of central Pacific El Niños has also been examined (Yu et al., 2012; Yeh et al., 2015; Ding et al., 2022). These studies provide a foundation for understanding the extratropical-tropical connection, highlighting the NPO's importance in ENSO development. However, the impact of climate change on these mechanisms remained an open question before this study.

This study utilized a set of large ensemble climate simulations from the Coupled Model Intercomparison Project Phase 6 (CMIP6). To determine the winter NPO pattern, the authors performed an EOF analysis on observed and detrended sea level pressure (SLP) anomalies over the North Pacific (20°-70°N, 120°E-100°W) from 1949-2021. The second EOF mode represented the NPO pattern. A common basis function approach was used to obtain the winter NPO index in both observations and CMIP6 models, projecting detrended winter SLP anomalies onto the observed EOF2. The NPO indices were standardized to have a standard deviation of one. Thirty-seven CMIP6 models were evaluated for their ability to simulate the observed winter NPO and ENSO seasonality. Models that accurately reproduced the observed relationship between winter NPO and subsequent winter ENSO (based on correlation between D-1JF0 NPO index and D0JF1 Niño3.4 SST index) were selected for further analysis (29 models). The correlation between the D-1JF0 NPO index and the D0JF1 Niño3.4 SST index was examined to evaluate model performance, removing the D-1JF0 ENSO signal to avoid signal overlap. The response of tropical SST and precipitation anomalies in OND0JF1 to the D-1JF0 NPO was compared between the present-day (1900-1990) and future (2007-2097) climate periods. A bootstrap test was used to assess the statistical significance of differences. The intensity of the WES feedback was described using the WES parameter (∂L/∂u - L/u), representing the change in surface latent heat flux (L) per unit change in surface wind (u). Numerical simulations using the Community Atmospheric Model version 5.0 of the Community Earth System Model version 1 were conducted to investigate the role of background SST changes in the enhanced WES feedback. Large ensemble experiments (100-member MPI-ESM and 50-member CanESM2) were used to further validate the findings. The intensity of the NPO's southern and northern lobes was quantified by averaging D-1JF0 SLP anomalies over specific regions. ENSO variability was represented by the boreal winter Niño3.4 SST index (5°S–5°N, 120°–170°W).

The study found a significant increase in the influence of the winter NPO on subsequent winter ENSO under global warming. This is evident in several key aspects: 1. **Enhanced Tropical Pacific Response:** Most CMIP6 models (25 of 29 selected, 86%) showed a strengthened response of tropical central-eastern Pacific (TCEP) SST and precipitation to preceding winter NPO in the future climate. The TCEP SST response increased by 32% on average, and precipitation response increased by 65%. This enhanced response was also observed in surface westerly wind anomalies and SST warming in the TCEP, further supporting the amplified influence of NPO on ENSO. 2. **Increased Frequency of NPO-ENSO Events:** The frequency of NPO events followed by ENSO events increased significantly in the future climate. This increase applies both to positive NPO events followed by El Niño and negative NPO events followed by La Niña. The increase in events was also observed for strong El Niño and La Niña events, strengthening the evidence for an amplified connection. 3. **Strengthened WES Feedback:** The wind-evaporation-SST (WES) feedback in the subtropical North Pacific is significantly enhanced in a warmer climate, attributable to a higher background mean SST and nonlinear SST-evaporation relationship. The enhanced WES feedback leads to stronger surface zonal wind anomalies over the tropical western-central Pacific (TWCP), thereby amplifying the impact of the NPO on ENSO. This is confirmed by additional numerical experiments using the Community Atmospheric Model version 5.0 of the Community Earth System Model version 1, which demonstrated that higher background SSTs lead to stronger surface wind responses in the tropical Pacific. 4. **Large Ensemble Confirmation:** The strengthened NPO-ENSO relationship in a warming climate was further validated using large ensemble simulations from MPI-ESM and CanESM2 models, independently confirming the CMIP6 findings. These experiments mirrored the results from CMIP6 models, highlighting a strengthened correlation between winter NPO and subsequent Niño3.4 SST index, and an increase in NPO-ENSO event frequency. They also confirmed that the WES feedback intensifies and the JJAO surface zonal wind response in the TWCP amplifies in the future climate. 5. **Role of Spring PMM:** This study also investigated the influence of the Spring Pacific Meridional Mode (PMM) as it is frequently associated with winter NPO, the findings show that the winter NPO's amplified influence on the subsequent spring PMM, further intensifies the influence on ENSO.

The study's findings directly address the research question concerning how global warming alters the NPO's influence on ENSO. The results highlight a clear strengthening of this connection, primarily due to the enhanced WES feedback under warmer SST conditions. The nonlinear relationship between SST and evaporation plays a key role in amplifying the effects of NPO-generated wind anomalies. The robustness of these findings is supported by both CMIP6 multi-model analysis and independent large ensemble simulations. This strengthening of the NPO-ENSO link carries significant implications for ENSO prediction, suggesting that incorporating winter NPO information could improve the skill of long-lead ENSO forecasts. The identified mechanism, based on strengthened WES feedback, provides a deeper understanding of how extratropical variability modulates ENSO under a changing climate. Further research should explore the influence of other extratropical modes and potential nonlinearities in the NPO-ENSO interaction.

This research demonstrates a significantly strengthened influence of the boreal winter North Pacific Oscillation (NPO) on El Niño-Southern Oscillation (ENSO) development in a warming climate. This amplification is primarily attributed to a more potent wind-evaporation-SST feedback in the subtropical North Pacific driven by higher background sea surface temperatures. This finding has important implications for improving long-lead ENSO prediction by leveraging the winter NPO as a predictor. Future research should investigate the role of other extratropical patterns and nonlinearities in this relationship under future climate scenarios.

While the study utilizes a large number of CMIP6 models and independent large ensemble simulations, there is still inherent uncertainty associated with climate model projections. The models may not perfectly capture all the complexities of the NPO-ENSO interaction. Furthermore, other extratropical factors influencing ENSO may not be fully considered, potentially limiting the study's comprehensive understanding. Despite the robustness of the findings, some individual models showed discrepancies in the NPO-ENSO connection under global warming, warranting further exploration into the potential for model-specific biases.