Takahito Mitsui

For the Internet of Things (IoT) comprising large-scale networks composed of tiny devices with limited resources, decentralized management is necessary to avoid an increase in load and cost from control signals; in addition, power-saving protocols such as intermittent communications are also essential. Carrier sense multiple access with collision avoidance (CSMA/CA) and time division multiple access (TDMA) have been used as one medium access control protocols in such environments, and CSMA/CA is one of the random access protocols which can be applied to the decentralized system; however, it causes significant delays by stochastic collisions in congested environments. Desynchronization is one of the concepts for decentralized TDMA, and the possibility of a collision can be reduced based on evenly distributed phases by interaction based on the phase differences. There are several desynchronization models, such as the Kuramoto-based model; however, these Kuramoto-based models have not been designed and implemented for IoT devices; it needs to exchange phase information continuously with others within sending cycles, thus, making it difficult to achieve ON/OFF scheduling in one cycle, intermittent communication. To make it applicable to IoT devices, we propose an intermittent inhibitory-coupled Kuramoto oscillator model in this paper. IoT devices can achieve autonomous scheduling to avoid collisions even in high-density environments by the proposed method; therefore, it can guarantee communication quality, moreover, be applied to power-saving IoT devices that use intermittent communication. Simulations and experiments have shown the effectiveness of the proposed method in the environments in both static networks and dynamical networks, supposing mobile environments. Furthermore, compared to CSMA/CA, the proposed method obtains a higher packet success rate than CSMA/CA.

While the Mid-Pleistocene Transition (MPT) is often described as a shift from 41-kyr to $\sim$100-kyr glacial cycles, this binary perspective fails to capture the nuanced spectral evolution of Quaternary climate. Applying wavelet-based spectral analysis to benthic d18O records, we identify previously underappreciated signals－~52 kyr before 1.2 Ma and ~76 kyr thereafter－marking the MPT's onset. Insolation analysis reveals these timescales primarily correspond to integer multiples (n) of the local-mean climatic precession period (T_p), with minor modulation by obliquity. We further find that when T_p falls within this intermediate range (20.5－22 kyr), high eccentricity triggers deglaciation, precluding the emergence of 60－65 kyr and 100－110 kyr cycles. These findings refine the conventional ``41-kyr to 100-kyr'' paradigm, revealing an evolution from a quasi-41-kyr regime including ~52-kyr cycles to a post-MPT regime of quantized ~76, ~95, and ~120~kyr cycles. This points to precession-based pacing as a persistent feature of Quaternary glacial dynamics.

Prediction and mitigation of extreme weather events are important scientific and societal challenges. Recently, Miyoshi and Sun (2022) proposed a control simulation experiment framework that assesses the controllability of chaotic systems under observational uncertainty, and within this framework, Sun et al. (2023) developed a method to prevent extreme events in the Lorenz 96 model. However, since their method is primarily designed to apply control inputs to all grid variables, the success rate decreases to approximately 60 % when applied to a single site, at least in a specific setting. Herein, we propose an approach that mitigates extreme events by updating local interventions based on multi-scenario ensemble forecasts. Our method achieves a high success rate, reaching 94 % even when applying interventions at one site per step, albeit with a moderate increase in the intervention cost. Furthermore, the success rate increases to 99.4 % for interventions at two sites. Unlike control-theoretic approaches adopting a top–down strategy, which determine inputs by optimizing cost functions, our bottom–up approach mitigates extreme events by effectively utilizing limited intervention options.

The dominant period of the Late Pleistocene glacial–interglacial cycles is roughly 100 kyr, rather than other major astronomical periods such as 19, 23, 41, and 400 kyr. Various models explain this fact through distinct dynamical mechanisms, including synchronization of self-sustained oscillations and resonance in mono- or multi-stable systems. However, the diversity of proposed models and dynamical mechanisms may obscure the essential factor behind the emergence of the ∼100 kyr periodicity. We propose the hypothesis that the ice-sheet climate system responds to astronomical forcing at the ∼100 kyr periodicity because the intrinsic timescale of the system is closer to 100 kyr than to other major astronomical periods. We support this idea with analyses and sensitivity studies of several simple ice age models with contrasting mechanisms.

Prediction and mitigation of extreme weather events are important scientific and societal challenges. Recently, Miyoshi and Sun (2022) proposed a control simulation experiment framework that assesses the controllability of chaotic systems under observational uncertainty, and within this framework, Sun et al. (2023) developed a method to prevent extreme events in the Lorenz 96 model. However, since their method is primarily designed to apply control inputs to all grid variables, the success rate decreases to approximately 60 % when applied to a single site, at least in a specific setting. Herein, we propose an approach that mitigates extreme events through local interventions based on multi-scenario ensemble forecasts. The success rate of our method is markedly higher than that of Sun et al.'s method, reaching 94 % even when applying interventions at one site per step, albeit with a moderate increase in the intervention cost. Furthermore, the success rate increases to 99.4 % during interventions at two sites. Unlike control-theoretic approaches adopting a top–down strategy, which determine inputs by optimizing cost functions, our bottom–up approach mitigates extreme events by effectively utilizing limited intervention options.

Prediction and mitigation of extreme weather events are important scientific and societal challenges. Recently, Miyoshi and Sun (2022) proposed a control simulation experiment framework that assesses the controllability of chaotic systems under observational uncertainty, and within this framework, Sun et al. (2023) developed a method to prevent extreme events in the Lorenz 96 model. However, since their method is primarily designed to apply control inputs to all grid variables, the success rate decreases to approximately 60 % when applied to a single site, at least in a specific setting. Herein, we propose an approach that mitigates extreme events through local interventions based on multi-scenario ensemble forecasts. The success rate of our method is markedly higher than that of Sun et al.'s method, reaching 94 % even when applying interventions at one site per step, albeit with a moderate increase in the intervention cost. Furthermore, the success rate increases to 99.4 % during interventions at two sites. Unlike control-theoretic approaches adopting a top–down strategy, which determine inputs by optimizing cost functions, our bottom–up approach mitigates extreme events by effectively utilizing limited intervention options.

Prediction and mitigation of extreme weather events are important scientific and societal challenges. Recently, Miyoshi and Sun (2022) proposed a control simulation experiment framework that assesses the controllability of chaotic systems under observational uncertainty, and within this framework, Sun et al. (2023) developed a method to prevent extreme events in the Lorenz 96 model. However, since their method is primarily designed to apply control inputs to all grid variables, the success rate decreases to approximately 60 % when applied to a single site, at least in a specific setting. Herein, we propose an approach that mitigates extreme events through local interventions based on multi-scenario ensemble forecasts. The success rate of our method is markedly higher than that of Sun et al.'s method, reaching 94 % even when applying interventions at one site per step, albeit with a moderate increase in the intervention cost. Furthermore, the success rate increases to 99.4 % during interventions at two sites. Unlike control-theoretic approaches adopting a top–down strategy, which determine inputs by optimizing cost functions, our bottom–up approach mitigates extreme events by effectively utilizing limited intervention options.

In recent years, the application of machine learning approaches to time-series forecasting of climate dynamical phenomena has become increasingly active. It is known that applying a bandpass filter to a time-series data is a key to obtaining a high-quality data-driven model. Here, to obtain longer-term predictability of machine learning models, we introduce a new type of bandpass filter. It can be applied to realtime operational prediction workflows since it relies solely on past time series. We combine the filter with reservoir computing, which is a machine-learning technique that employs a data-driven dynamical system. As an application, we predict the multi-year dynamics of the El Niño-Southern Oscillation with the prediction horizon of 24 months using only past time series.

Prediction and mitigation of extreme weather events are important scientific and societal challenges. Recently, Miyoshi and Sun (2022) proposed a control simulation experiment framework that assesses the controllability of chaotic systems under observational uncertainty, and within this framework, Sun et al. (2023) developed a method to prevent extreme events in the Lorenz 96 model. However, since their method is primarily designed to apply control inputs to all grid variables, the success rate decreases to approximately 60 % when applied to a single site, at least in a specific setting. Herein, we propose an approach that mitigates extreme events through local interventions based on multi-scenario ensemble forecasts. The success rate of our method is markedly higher than that of Sun et al.'s method, reaching 94 % even when applying interventions at one site per step, albeit with a moderate increase in the intervention cost. Furthermore, the success rate increases to 99.4 % during interventions at two sites. Unlike control-theoretic approaches adopting a top–down strategy, which determine inputs by optimizing cost functions, our bottom–up approach mitigates extreme events by effectively utilizing limited intervention options.

Millennial-scale climate variability during Pleistocene Ice Ages, known as Dansgaard-Oeschger (DO) cycles, are characterised by abrupt transitions between Greenland cold stadials and warm interstadials, which coincide with gradual warming and cooling over Antarctica, respectively, via the bipolar seesaw. DO cycles are associated with reorganisations of the Atlantic Meridional Overturning Circulation (AMOC), but the mechanisms driving them remain unclear. In this study, from nudging experiments based on intrinsic millennial-scale AMOC variability in a complex climate model, we show that gradual changes in sea ice over the Southern Ocean induced by the bipolar seesaw act as a negative feedback to maintain the millennial-scale AMOC variability. Southern Ocean surface cooling during the interstadial phases enhances regional sea ice-related salt and freshwater fluxes, which eventually weakens the AMOC by strengthening the oceanic stratification over the North Atlantic by increasing and decreasing the salinity of Antarctic bottom water and Antarctic intermediate water, respectively. The Southern Ocean feedback becomes particularly important for DO cycles with long periodicities, such as those occurring during Marine Isotope Stages 5, 4, 2 and those appearing after major Heinrich events. Our results suggest that the Southern Ocean feedback helps drive the DO cycles, demonstrating the globally connected nature of these events.