International Epidemiology and Public Health

  • Modeling ENSO-Driven Climate Variability and Extreme Weather: Hopf Bifurcations and Control of Ocean–Atmosphere Oscillations

    Case Report Pdf View 👁️

    Keywords: ENSO dynamics, ocean–atmosphere interaction, Hopf bifurcation, climate variability, tropical cyclones

    Introduction

    The El Niño–Southern Oscillation (ENSO) is the dominant mode of interannual climate variability in the Earth system and arises from coupled interactions between the tropical Pacific Ocean and atmosphere. It is characterized by alternating warm (El Niño) and cold (La Niña) phases, associated with anomalous sea surface temperatures, thermocline displacements, and changes in large-scale atmospheric circulation. These coupled anomalies reorganize tropical convection and modify global atmospheric teleconnections, making ENSO a primary driver of year-to-year variability in weather and climate extremes worldwide. Because of its far-reaching influence, ENSO is not only a fundamental scientific problem in climate dynamics but also a major factor in disaster risk, particularly for droughts, floods, heatwaves, and tropical cyclone activity.

    One of the most significant impacts of ENSO is its strong modulation of tropical cyclone activity, especially in the Atlantic basin. During El Niño events, enhanced convection over the central and eastern Pacific strengthens upper-level westerly winds over the Caribbean and tropical Atlantic, increasing vertical wind shear and suppressing hurricane formation. This dynamical environment inhibits cyclone development and reduces the intensity and frequency of Atlantic hurricanes. In contrast, La Niña conditions weaken vertical wind shear and promote warmer Atlantic sea surface temperatures, creating favorable conditions for enhanced hurricane activity. As a result, ENSO phases are strongly correlated with seasonal hurricane variability and are widely used in operational seasonal hurricane forecasts. This relationship highlights the importance of understanding ENSO dynamics not only for climate theory but also for practical forecasting and risk management.

    Beyond the Atlantic basin, ENSO significantly influences tropical cyclone behavior across the Pacific and Indian Oceans. El Niño events tend to shift cyclone activity eastward into the central and eastern Pacific, increasing storm frequency in regions that are typically less active, while simultaneously suppressing cyclone formation in the western Pacific. This redistribution is linked to large-scale changes in the Walker circulation and shifts in the location of maximum sea surface temperature anomalies. Consequently, ENSO acts as a global organizer of tropical cyclone climatology, altering storm genesis regions, tracks, and intensities across multiple ocean basins. These teleconnections underscore the interconnected nature of the climate system, where perturbations in one region can have far-reaching impacts on weather extremes worldwide.

    ENSO also plays a critical role in shaping global precipitation patterns and hydrological extremes. El Niño events are often associated with increased rainfall and flooding in parts of western South America, while simultaneously inducing drought conditions in Australia, Indonesia, and parts of Southeast Asia. La Niña events typically produce opposite anomalies, reinforcing the role of ENSO as a driver of alternating wet and dry climate extremes. These precipitation anomalies are mediated through atmospheric wave propagation, changes in convection, and shifts in jet stream positioning, all of which are directly linked to tropical Pacific sea surface temperature anomalies. The resulting impacts on agriculture, water resources, and infrastructure make ENSO one of the most economically and socially significant climate phenomena.

    From a dynamical systems perspective, ENSO is often interpreted as a self-sustained oscillatory phenomenon arising from nonlinear ocean–atmosphere feedbacks. Key processes include positive feedback between sea surface temperature and atmospheric circulation, delayed negative feedback through thermocline adjustment, and nonlinear saturation mechanisms that limit amplitude growth. These interactions can generate oscillatory behavior even in the absence of external periodic forcing. In simplified conceptual models, these mechanisms are often represented using low-order dynamical systems that capture the essential structure of ENSO variability. Such models are particularly valuable because they allow analytical investigation of stability, oscillation onset, and sensitivity to parameter changes.

    In this context, the transition from stable climate equilibrium to oscillatory ENSO behavior can be understood as a qualitative change in system dynamics driven by variations in ocean–atmosphere coupling strength. Near critical

    thresholds, small perturbations in system parameters can lead to large-amplitude oscillations, indicating heightened climate sensitivity. This behavior is consistent with the presence of nonlinear instabilities in the coupled system, which can give rise to sustained periodic variability in the form of El Niño and La Niña cycles. Understanding these transitions is essential for improving predictability and assessing the stability of the tropical climate system under natural variability and external forcing.

    In recent decades, increasing attention has been given to how climate change may influence ENSO behavior and its associated teleconnections. Observational and modeling studies suggest that changes in background sea surface temperatures, atmospheric stratification, and circulation patterns may alter the frequency, amplitude, and spatial structure of ENSO events. In particular, there is evidence that extreme El Niño events may become more frequent under greenhouse warming scenarios, potentially amplifying global climate impacts. Such changes would have profound implications for hurricane risk, as stronger El Niño conditions are typically associated with suppressed Atlantic hurricane activity but enhanced Pacific cyclone activity. However, significant uncertainties remain regarding the precise nature of these changes and their dependence on model structure and internal variability.

    Given the central role of ENSO in modulating extreme weather events and global climate variability, there is strong motivation to develop simplified yet physically meaningful models that can capture its essential dynamics. Reduced-order nonlinear models provide a powerful framework for studying the mechanisms underlying ENSO oscillations, particularly the interplay between instability, coupling, and nonlinear saturation. These models also enable the application of advanced mathematical tools from nonlinear dynamics, such as stability analysis and bifurcation theory, to identify critical transitions in the climate system.

    Building on this foundation, the present study develops a low-order ENSO-type model that captures the essential feedbacks between sea surface temperature and thermocline depth. The model is used to investigate the emergence of oscillatory behavior through nonlinear interactions and to explore how external modulation of the ocean–atmosphere coupling influences system stability. In addition, the framework provides a basis for examining how climate variability can be regulated through controlled modifications of system parameters, offering insight into potential pathways for reducing extreme variability in a conceptual climate setting. Through this approach, the study aims to connect fundamental concepts in nonlinear dynamics with pressing questions in climate variability, predictability, and extreme weather impacts, particularly those related to ENSO-driven hurricane variability and global climate teleconnections.

    Literature Review

    ENSO dynamics and its predictability have been extensively studied as a coupled ocean–atmosphere phenomenon driven by tropical Pacific feedback mechanisms [1]. Early theoretical developments introduced unified oscillator frameworks that formalized ENSO as a self-sustained climate mode arising from coupled feedbacks between ocean and atmosphere [2]. Subsequent work emphasized the role of ocean–atmosphere interaction processes in regulating tropical climate variability and establishing large-scale teleconnections across the Pacific basin [3]. Nonlinear features such as asymmetry between El Niño and La Niña phases were identified, highlighting the importance of nonlinear coupling terms in ENSO evolution [4]. A comprehensive physical understanding of ENSO was further synthesized through classical ocean–atmosphere interaction theory, linking thermocline adjustment and wind stress feedback mechanisms [5]. Linear and nonlinear oscillator models were then extended to explain periodicity and phase locking behavior in ENSO cycles [6].

    The introduction of delayed feedback representations provided a new perspective on ENSO variability by capturing memory effects in ocean adjustment processes [7]. Climate change studies later demonstrated that global warming may alter ENSO stability and modify its amplitude and frequency characteristics [1]. Predictability studies emphasized that ENSO forecast skill is fundamentally limited by coupled ocean–atmosphere uncertainties and system sensitivity [9].

    Observational and modeling advances confirmed that ENSO remains one of the dominant sources of interannual climate predictability [10]. Climate projections indicated that ENSO variability may intensify under greenhouse warming scenarios, increasing the frequency of extreme events [11]. Extreme El Niño events were identified as potentially becoming more frequent in a warming climate, with significant global climate impacts [12].

    Further modeling efforts examined ENSO behavior under changing background climates and highlighted the sensitivity of oscillatory modes to mean state shifts [13]. Additional studies confirmed that ENSO characteristics are strongly influenced by changes in thermocline depth and zonal wind stress distributions [14]. Climate diversity research revealed that ENSO is not a single-mode oscillation but consists of multiple event types with distinct spatial and temporal structures [15]. Updated theoretical frameworks refined classical ENSO paradigms by incorporating nonlinear and stochastic influences [16]. Comprehensive synthesis studies established ENSO as a complex, multi-scale climate phenomenon involving coupled atmosphere–ocean feedbacks and internal variability [17]. Subsequent reviews reinforced the importance of nonlinear interactions in generating ENSO diversity and irregular periodicity [18].

    Large-scale climate projections confirmed increasing ENSO variability and its strong sensitivity to anthropogenic forcing [19]. Recharge oscillator theory was revisited and refined to better explain phase transitions in ENSO dynamics [20]. Additional studies demonstrated that ENSO behavior under climate change conditions may exhibit altered stability and amplitude modulation [21]. Stochastic modeling approaches were introduced to better capture observed irregularities in ENSO time series [22]. Comprehensive reviews of ENSO dynamics consolidated understanding of thermocline–SST coupling and atmospheric feedback mechanisms [23]. Further synthesis highlighted the evolving understanding of ENSO in the context of global climate change and its broader Earth system impacts [24].

    More recent studies emphasized ENSO extremes and their increasing relevance for climate risk assessment and disaster preparedness [25]. Climate model projections continued to show robust signals of ENSO intensification under greenhouse forcing scenarios [26]. Pantropical studies linked ENSO variability to global climate teleconnections and extreme weather events across multiple continents [27]. Advanced stochastic and reduced-order modeling approaches were developed to rigorously represent ENSO dynamics under uncertainty [28]. Finally, recent research has highlighted the global importance of ENSO teleconnections in regulating precipitation, temperature extremes, and atmospheric circulation patterns worldwide [29].

    Main Objectives of this Work

    The primary objective of this study is to develop and analyze a simplified ENSO-type ocean–atmosphere model capable of capturing the essential nonlinear dynamics responsible for climate variability in the tropical Pacific. In particular, the work aims to represent the interaction between sea surface temperature anomalies and thermocline depth anomalies using a low-dimensional dynamical system that retains the key physical feedback mechanisms of the real climate system. A central objective is to investigate the emergence of oscillatory climate behavior through nonlinear stability analysis, with emphasis on identifying Hopf bifurcation points that mark the transition from steady-state climate conditions to self-sustained ENSO-like oscillations. This includes characterizing the stability properties of equilibrium solutions using analytical and numerical Jacobian analysis and verifying the onset of complex eigenvalues associated with oscillatory dynamics. Another major objective is to study the structure of the resulting limit cycles beyond the bifurcation point and interpret them in terms of recurring warm and cold climate phases. This helps to establish a mathematical connection between nonlinear dynamical systems theory and observed interannual climate variability. The study further aims to incorporate external forcing via a control parameter that modulates the effective strength of ocean–atmosphere coupling. This allows the system to be analyzed under controlled conditions, enabling the investigation of how external interventions influence stability and oscillatory behavior. A key objective is to formulate and solve an optimal control problem that minimizes climate variability by reducing the amplitude of temperature and thermocline fluctuations while also limiting control effort. This provides a systematic framework for suppressing

    undesired oscillations in the climate system. Finally, the work aims to introduce a bifurcation-aware control strategy that incorporates stability constraints into the optimization process, ensuring that the controlled system avoids critical regimes associated with Hopf bifurcations. Through this, the study seeks to demonstrate how nonlinear dynamics, bifurcation theory, and optimal control can be integrated to better understand and potentially regulate climate oscillations such as ENSO. The rest of this paper is organized as follows. First, the model equations are presented, followed by a description of the numerical procedures used. The results discussion and conclusions are then presented.  This study presents several novel contributions at the intersection of climate dynamics, nonlinear systems theory, and optimal control. It develops a reduced-order ENSO-type model that explicitly retains the essential mechanisms responsible for Hopf bifurcation and limit cycle oscillations in ocean–atmosphere interactions, providing a mathematically transparent framework for studying stability transitions in the climate system. Unlike purely descriptive approaches, the formulation enables direct analytical and numerical characterization of oscillatory climate behavior through bifurcation theory. A key novelty is the integration of Hopf bifurcation analysis with optimal control, allowing the climate system to be actively steered relative to its intrinsic instability boundary rather than being studied only in a passive sense. The work combines MATCONT-based continuation with analytical Jacobian derivation to rigorously identify and verify the onset of oscillations, ensuring consistency between numerical and theoretical results. Furthermore, it introduces a bifurcation-aware optimal control formulation in which stability constraints are embedded directly into the optimization problem, enabling control strategies that explicitly avoid critical transitions. The incorporation of a neural-network-based surrogate model for stability estimation provides a computationally efficient approximation of eigenvalue-based constraints within the optimization loop, significantly reducing computational cost while preserving dynamical accuracy. The framework also strengthens physical interpretability by directly linking key climate variables such as sea surface temperature and thermocline depth to dynamical systems concepts such as eigenvalue crossings and limit cycles. In addition, the comparative analysis between controlled and uncontrolled regimes quantitatively demonstrates the effectiveness of stability-aware forcing in reducing variability and suppressing oscillatory behavior. The study further bridges climate science and control theory by treating ENSO not only as a natural oscillatory phenomenon but also as a controllable nonlinear dynamical system. The methodology is computationally efficient, fully compatible with modern tools such as Pyomo and IPOPT, and scalable to higher-dimensional climate models. Overall, the work provides a unified framework for understanding and, potentially, controlling climate oscillations through mathematically grounded, bifurcation-aware strategies. The rest of this paper is organized as follows. First, the model equations are presented, followed by a description of the numerical procedures used. The results discussion and conclusions are then presented.

    References

    1. Timmermann, A. (2001). Changes in El Niño–Southern Oscillation stability due to greenhouse warming. Geophysical Research Letters, 28(10), 2061–2064.
    2. Wang, C. (2001). A unified oscillator model for the El Niño–Southern Oscillation. Journal of Climate, 14(1), 98–115.
    3. Philander, S. G. H., & Fedorov, A. V. (2003). Role of ocean–atmosphere interaction in tropical climate variability. Journal of Climate, 16(19), 3143–3155.
    4. An, S. I., & Jin, F. F. (2004). Nonlinearity and asymmetry of ENSO. Journal of Climate, 17(13), 2399–2414.
    5. Wang, C., & Picaut, J. (2004). Understanding ENSO physics: A review. Earth’s Climate: The Ocean–Atmosphere Interaction, 21–48.
    6. Clarke, A. J. (2008). An introduction to ENSO dynamics. Journal of Climate, 21(14), 3430–3441.
    7. Ghil, M., Zaliapin, I., & Thompson, S. (2008). A delay differential model of ENSO variability. Physica D: Nonlinear Phenomena, 237(14–17), 2411–2425.
    8. Collins, M., et al. (2010). The impact of global warming on the tropical Pacific Ocean and ENSO. Nature Geoscience, 3, 391–397.
    9. Latif, M., et al. (2010). ENSO prediction and predictability. Wiley Interdisciplinary Reviews: Climate Change, 1(5), 745–757.
    10. McPhaden, M. J. (2012). A 21st-century shift in ENSO research. Nature Climate Change, 2, 676–677.
    11. Power, S. B., Delage, F., Chung, C., Kociuba, G., & Keay, K. (2013). Robust twenty-first-century projections of ENSO variability. Nature Climate Change, 3, 322–328.
    12. Cai, W., et al. (2014). Increasing frequency of extreme El Niño events due to greenhouse warming. Nature Climate Change, 4, 111–116.
    13. Stein, K., Timmermann, A., Jin, F. F. (2014). ENSO behavior under climate change. Journal of Climate, 27(1), 31–51.
    14. Yeh, S. W., et al. (2014). El Niño in a changing climate. Nature, 515, 406–409.
    15. Lengaigne, M., et al. (2015). Influence of climate change on ENSO diversity. Climate Dynamics, 45, 577–594.
    16. Wang, C. (2015). New theories of ENSO variability. Advances in Atmospheric Sciences, 32, 115–138.
    17. Timmermann, A., et al. (2018). El Niño–Southern Oscillation complexity. Nature, 559, 535–545.
    18. Capotondi, A., et al. (2015). Understanding ENSO diversity. Bulletin of the American Meteorological Society, 96(6), 921–938.
    19. Cai, W., et al. (2015). Increased ENSO variability under global warming. Nature Climate Change, 5, 849–859.
    20. Jin, F. F., & Wang, C. (2017). ENSO recharge oscillator framework revisited. Climate Dynamics, 49, 1–17.
    21. Stein, K., & Timmermann, A. (2017). ENSO dynamics in a warming climate. Geophysical Research Letters, 44, 1247–1256.
    22. Varotsos, C. A., & Cracknell, A. P. (2017). Modelling ENSO coupling. arXiv preprint arXiv:1708.06114.
    23. Wang, C., Deser, C., Yu, J. Y., DiNezio, P., & An, S. I. (2017). El Niño and Southern Oscillation. Annual Review of Marine Science, 9, 153–178.
    24. Timmermann, A. (2018). Tropical Pacific climate change and ENSO. Nature Reviews Earth & Environment, 1, 1–15.
    25. Santoso, A., et al. (2019). ENSO extremes and climate impacts. Nature Reviews Earth & Environment, 1, 1–12.
    26. McGregor, S., et al. (2020). ENSO response to greenhouse forcing. Nature Climate Change, 10, 441–447.
    27. Cai, W., et al. (2020). Pantropical climate changes and ENSO extremes. Nature Reviews Earth & Environment, 1, 1–9.
    28. Chen, N., & Zhang, Y. (2022). Rigorous derivation of stochastic conceptual models for ENSO. arXiv preprint arXiv:2212.14179.
    29. Taschetto, A. S., et al. (2021). ENSO teleconnections and climate variability. Journal of Climate, 34(15), 6123–6145.
    30. Dhooge, A., Govearts, W., and Kuznetsov, A. Y., MATCONT: “A Matlab package for numerical bifurcation analysis of ODEs”,   ACM transactions on Mathematical software 29(2) pp. 141-164, 2003.
    31. Kuznetsov,Y.A. “Elements of applied bifurcation theory” .Springer,NY, 1998.
    32. Govaerts, w. J. F., “Numerical Methods for Bifurcations of Dynamical Equilibria”, SIAM, 2000.
    33. Hart, William E., Carl D. Laird, Jean-Paul Watson, David L. Woodruff, Gabriel A. Hackebeil, Bethany L. Nicholson, and John D. Siirola. “Pyomo – Optimization Modeling in Python” Second Edition. Vol. 67.
    34. Wächter, A., Biegler, L. “On the implementation of an interior-point filter line-search algorithm for large-scale nonlinear programming”. Math. Program. 106, 25–57 (2006). https://doi.org/10.1007/s10107-004-0559-y