The evaporation duct (ED) is a special type of atmospheric layer that can trap electromagnetic waves of certain frequencies, enabling over-the-horizon transmission. Accurate diagnosis and prediction of ED height (EDH) have a significant impact on the performance of the radar and communication equipment. This article proposes a hybrid temporal convolutional network-gated recurrent unit (TCN-GRU) EDH diagnostic model that incorporates the height parameters and reconstructs field campaign data distributions by random sampling. It provides a new research perspective and supplements the shortcomings of conventional approaches, which rely solely on time-series data. The model's performance is evaluated through comparative experiments with traditional physical theory models (Babin-Young-Carton (BYC), Musson-Genon-Bruth (MGB), naval postgraduate school (NPS), and LiuLi2.0) and artificial intelligence models (extreme gradient boosting (XGB) and LSTM) across multiple field campaign datasets. The results demonstrate that the proposed TCN-GRU EDH diagnostic model has improved EDH diagnostic accuracy.

Accurate long-range forecasting of the El -Southern Oscillation (ENSO) is vital for global climate prediction and disaster risk management. Yet, limited understanding of ENSO's physical mechanisms constrains both numerical and deep learning approaches, which often struggle to balance predictive accuracy with physical interpretability. Here, we introduce a data driven model for ENSO prediction based on conditional diffusion model. By constructing a probabilistic mapping from historical to future states using higher-order Markov chain, our model explicitly quantifies intrinsic uncertainty. The approach achieves extending lead times of state-of-the-art methods, resolving early development signals of the spring predictability barrier, and faithfully reproducing the spatiotemporal evolution of historical extreme events. The most striking implication is that our analysis reveals that the reverse diffusion process inherently encodes the classical recharge-discharge mechanism, with its operational dynamics exhibiting remarkable consistency with the governing principles of the van der Pol oscillator equation. These findings establish diffusion models as a new paradigm for ENSO forecasting, offering not only superior probabilistic skill but also a physically grounded theoretical framework that bridges data-driven prediction with deterministic dynamical systems, thereby advancing the study of complex geophysical processes.

Data‐driven weather prediction (DDWP) has made significant advancements in recent years. However, weather prediction using DDWPs still requires an accurate initial field as the input. To fulfill this requirement, the four‐dimensional variational (4DVar) approach can offer initial fields. Recent studies have demonstrated the potential of deep learning (DL)‐based methods in accelerating 4DVar. In this study, we propose a novel model called the 4DVar‐informed generative adversarial network (4DVarGAN), which combines prior knowledge from 4DVar with the conditional generative network (CGAN). We employ a CGAN to non‐iteratively solve the 4DVar cost function and utilize a cycle‐consistent adversarial learning framework for data augmentation. Additionally, we incorporate a 4DVar‐based adaptive adjustment to the output of the proposed model's analysis increment‐generating component, which promotes reasonable stabilization. Experimental results using 500 hPa geopotential fields from the WeatherBench data set demonstrate that our approach achieves a 73‐fold acceleration compared to the 4DVar implemented by the DDWP model. Furthermore, our model exhibits the lowest initial and forecast errors, outperforming state‐of‐the‐art DL‐based data assimilation (DA) methods. Moreover, our method demonstrates effective performance when starting from background fields of varying qualities, consistently achieving stable results. These findings highlight the potential of CGANs in enhancing the reliability of data‐driven DA by incorporating the prior knowledge of the 4DVar method.

Atmospheric General Circulation Model (AGCM) is widely used for predicting climate change and extreme weather. AGCM has been scaled up to thousands of cores on HPC platforms to provide timely forecasts. However, the communication overhead significantly impacts the performance and scalability of AGCM when using a two-dimensional domain decomposition algorithm in parallel. To tackle this issue, we introduce Pipe-AGCM, a fine-grain pipelining scheme to reduce communication overhead. We divide latitudes into groups to overlap communication from one group with the computation of another. By adjusting the group count, we can establish a pipelining scheme, optimizing AGCM performance. Pipe-AGCM has demonstrated efficiency and scalability in experiments on HPC platforms. The pipelining scheme reduced communication time and wall clock time by 78.02

In the earth sciences, numerical weather prediction (NWP) is the primary method of predicting future weather conditions, and its accuracy is affected by the initial conditions. Data assimilation (DA) can provide high-precision initial conditions for NWP. The hybrid 4DVar-EnKF is currently an advanced DA method used by many operational NWP centres. However, it has two major shortcomings: The complex development and maintenance of the tangent linear and adjoint models and the empirical combination of the results of 4DVar and EnKF. In this paper, a new hybrid DA method based on machine learning (HDA-ML) is presented to overcome these drawbacks. In the new method, the tangent linear and adjoint models in the 4DVar part of the hybrid algorithm can be easily obtained by using a bilinear neural network to replace the forecast model, and a CNN model is adopted to fuse the analysis of 4DVar and EnKF to adaptively obtain the optimal coefficient of combination rather than the empirical coefficient as in the traditional hybrid DA method. The hybrid DA methods are compared with the Lorenz-96 model using the true values as labels. The experimental results show that HDA-ML improves the assimilation performance and significantly reduces the time cost. Furthermore, using observations instead of the true values as labels in the training system is more realistic. The results show comparable assimilation performance to that in the experiments with the true values used as the labels. The experimental results show that the new method has great potential for application to operational NWP systems.

In coastal cities, accurate wave forecasting provides vital safety for the marine operations of ships and the construction of coastal projects. However, it is challenging to accurately forecast ocean waves due to their non-linear and non-smooth characteristics. To overcome this difficulty, we propose the ISP-FESAN method, which optimizes significant wave height prediction by feature engineering and self-attention networks. Specifically, in the process of feature engineering, we first perform the empirical modal decomposition (EMD) of the wave signal, then we add the decomposed sub-signals to the original dataset for the feature enhancement, and finally, we perform the feature selection on the dataset to determine the final input features for the self-attention network. Extensive experiments are conducted to verify the effectiveness of our method on 24-h and 48-h predictions. The results show that ISP-FESAN outperforms the other methods compared in our experiments.

The prediction of chaotic time series systems has remained a challenging problem in recent decades. A hybrid method using Hankel Alternative View Of Koopman (HAVOK) analysis and machine learning (HAVOK-ML) is developed to predict chaotic time series. HAVOK-ML simulates the time series by reconstructing a closed linear model so as to achieve the purpose of prediction. It decomposes chaotic dynamics into intermittently forced linear systems by HAVOK analysis and estimates the external intermittently forcing term using machine learning. The prediction performance evaluations confirm that the proposed method has superior forecasting skills compared with existing prediction methods.

The initial field has a crucial influence on numerical weather prediction (NWP). Data assimilation (DA) is a reliable method to obtain the initial field of the forecast model. At the same time, data are the carriers of information. Observational data are a concrete representation of information. DA is also the process of sorting observation data, during which entropy gradually decreases. Four-dimensional variational assimilation (4D-Var) is the most popular approach. However, due to the complexity of the physical model, the tangent linear and adjoint models, and other processes, the realization of a 4D-Var system is complicated, and the computational efficiency is expensive. Machine learning (ML) is a method of gaining simulation results by training a large amount of data. It achieves remarkable success in various applications, and operational NWP and DA are no exception. In this work, we synthesize insights and techniques from previous studies to design a pure data-driven 4D-Var implementation framework named ML-4DVAR based on the bilinear neural network (BNN). The framework replaces the traditional physical model with the BNN model for prediction. Moreover, it directly makes use of the ML model obtained from the simulation data to implement the primary process of 4D-Var, including the realization of the short-term forecast process and the tangent linear and adjoint models. We test a strong-constraint 4D-Var system with the Lorenz-96 model, and we compared the traditional 4D-Var system with ML-4DVAR. The experimental results demonstrate that the ML-4DVAR framework can achieve better assimilation results and significantly improve computational efficiency.

SummaryMany cyber‐attach schemes and coding models established by algebra tools are build to address the problem of security of cyber‐pysical systems (CPS). As an important field of algebra computing, Boolean Polynomial System Solving (PoSSo) problem plays a very important role in many algebra applications. In this article, we propose an efficient Parallel Boolean Characteristic Set method (PBCS) under the high‐performance computing environment to improve the efficiency of solving Boolean polynomial systems. The PBCS is implemented based on the state‐of‐the‐art Boolean Characteristic Set method (BCS). It adopts a master‐slave parallel pattern, and distributes tasks based on the polynomial sets after initial zero decomposition. We design a strategy of dynamically reallocating tasks to ameliorate load imbalance, which is caused by dynamical zero decomposition of polynomials. Furthermore, we improve its performance by optimizing the parameter settings of PBCS, including the maximum number of polynomial branches that trigger the dynamic allocation policy and the scheduling time. Experimental results with solving several Boolean polynomial systems confirm that PBCS is efficient and scalable, especially for the equations generating from stream ciphers that have block triangular structure. Moreover, the method also has good scalability. It shows a stable speedup as well even extending to the size of thousands of CPU cores.

Evaporation duct is an abnormal refraction atmospheric stratification frequently appearing at sea. Accurately obtaining evaporation duct height (EDH) is significant for the effective application of electromagnetic system equipment. To overcome the limitations of theoretical prediction models, this study proposes a pure data-driven back propagation neural network (BPNN) EDH prediction model (referred to as BPNN model) for the first time, which uses BPNN suitable for the data characteristics of evaporation duct and then conducts two groups of experiments to fully test the proposed BPNN model. The Paulus–Jeske (PJ) model and support vector regression (SVR) model are introduced as the baseline methods in these experiments. The results of the first group of experiments prove that on the EDH prediction in all the experimental areas, the BPNN model has significantly better comprehensive performances than the PJ model and the SVR model. Also, the second group of experiments’ results reflects the BPNN model has great generalisation capacity, which makes it suitable for predicting the EDH, and the distributions of EDH have a strong spatial correlation.

Serial ensemble Kalman filter (EnKF) is a kind of EnKF which treats observations serially during every assimilation step. The assimilation order can be generated by different rules and has significant impacts on the performance of serial EnKF when localization algorithm is applied. In this study, we seek to examine and better understand the characteristics of various ordering methods when they are applied in the serial EnKF. The results show that different ordering methods demonstrate almost the same changes in analysis as the localization radius changing. Moreover, the optimal parameters of localization radius and forgetting factor of serial EnKF are found varying among different ordering rules. In addition, a novel rule for confirming the assimilation order is proposed to further improve the performance of serial EnKF. The observations are sorted from “better” to “worse” (OBS-BtoW), which are evaluated by estimating the distance of analysis between the prior and observations. Compared with the existing ordering methods, the proposed method can improve the performance at a very small computation cost without needing future forecasts and the truth.

In this paper, we study the dynamics property of a stochastic HIV model with Beddington–DeAngelis functional response. It has a unique uninfected steady state. We prove that the model has a unique global positive solution. Furthermore, if the basic reproductive number is not larger than 1, the asymptotic behavior of the solution is stochastically stable. Otherwise, it fluctuates randomly around the infected steady state of the corresponding deterministic HIV model. Finally, some numerical simulations are carried out to verify our results.

The quality and accuracy of Numerical Weather Prediction (NWP) is based on its initial conditions (ICs), boundary conditions and forecast models. Data assimilation (DA) is a crucial procedure to optimally estimate the actual atmospheric state (known as the analysis field) as ICs for NWP by integrating available information, including the observation and the background field. Instead of only focusing on the speed-up for DA in virtue of the customized neural networks, this paper exploratively introduces the spatial-temporal peculiarities to construct a new hybrid data assimilation approach based on multilayer perceptron (MLP); and, its effectiveness and validity are verified in two classical nonlinear dynamic models. The results of experiments demonstrate that the Cache-MLP generally produces similar or smaller root mean square errors (RMSE) with much less time consuming, compared to the conventional 3D-Var and EnKF DA methods, and noticeably, the Cache-MLP has a more robust representation of turning points in the trajectories of the state variables. The final Backtracked-MLP learns a propriate weight matrix to couple previous two traditional DA methods and increases the accuracy by 10.32% in the Lorenz-63 system while 14.03% in the Lorenz-96 system, in comparison with the empirical hybrid DA method. To some extent, this method could be a reference to further researches to optimize the quality of the analysis field, in the meantime, saving significant computing time and resources by deep learning.

SummaryThe Atmospheric General Circulation Model (AGCM) as one of the most important components of Climate System Model (CSM), has been proved to be an effective way for weather forecasting and climate prediction. Although lots of efforts have been conducted to improve the computing efficiency of AGCMs, such as exploit parallel algorithms, migrating codes, and even redesigning systems to adapt to the emerging computer architectures, it is not enough to match the real requirement, due to the limited scalability of the parallel algorithms themselves. Therefore, we design and implement a scalable parallel spectral‐based atmospheric circulation mode called PAGCM in this paper. Specifically, we first analyze the data dependencies of the dimensions in different spaces according to the calculation characteristics of spectral models, and based on which we propose a two‐dimensional decomposition algorithm in PAGCM to effectively increase the involving cores for the parallel computing, and thus reduce the overall computing time. Furthermore, to adapt to the novel data decomposition in each computing stage of dynamic framework, we propose three‐dimensional data transposition algorithms and data collection algorithms correspondingly, by considering of load balancing and communication optimization. Extensive experiments are conducted on Tianhe‐2 to validate the effectiveness and scalability of our proposals.

Variations in both symmetric wind components and asymmetric wave amplitudes of a tropical cyclone depend on the location of its center. Because the radial structure of asymmetries is critical to the wave–mean interaction, this study, under idealized conditions, examines the influences of a center location on the radial structure of the diagnosed asymmetries. It has been found that the amplitudes of aliasing asymmetries are mainly affected by the initial symmetric fields. Meanwhile, the radial structure of asymmetry is controlled by the aliasing direction. Sensitivity tests on the location of the center were employed to emphasize the importance of the aliasing direction using angular momentum equations. With a small displacement, the tendencies of azimuthal tangential wind are found to reverse completely when the center shifts to a different direction. This work concludes that the diagnostic results related to asymmetric decomposition should be treated rigorously, as they are prone to inaccuracies, which in turn affect cyclone prediction.

Precise center-detection of tropical cyclones (TCs) is critical for dynamic analysis in high resolution model data. The existence of both smaller scale perturbations and larger scale circulations could reduce the accuracy of center positioning. In this study, an objective center-finding algorithm is developed based on a two-dimensional Fourier filter and a vorticity centroid algorithm. This proposed algorithm is able to automatically adjust its parameters according to the scale of the target vortex instead of using artificially prescribed parameters in previous research. What’s more, this new algorithm has been optimized and validated by a hundred idealized vortexes with different sizes and small-scale perturbations. A high-resolution simulation of Typhoon Soudelor (2015) was used to evaluate the performance of the new algorithm, and the proposed objective center-finding algorithm was found able to detect a precise and reliable center.

Solving Boolean polynomial systems as an important aspect of symbolic computation, plays a fundamental role in various real applications. Although there exist many efficient sequential algorithms for solving Boolean polynomial systems, they are inefficient or even unavailable when the problem scale becomes large, due to the computational complexity of the problem and the limited processing capability of a single node. In this paper we propose an efficient parallel characteristic set method called PBCS for solving Boolean polynomial systems under the high-performance computing environment. Specifically, PBCS takes full advantage of the state-of-the-art characteristic set method and achieves load balancing by dynamically reallocating tasks. Moreover, the performance is further improved by optimizing the parameter setting. Extensive experiments are conducted to demonstrate that PBCS is efficient and scalable for solving Boolean equations, especially for the equations rasing from stream ciphers that have block triangular structure. In addition, the algorithm has good scalability and can be extended to the size of thousands CPU cores with a stable speedup.

In large-scale open distributed environment such as Grid computing, Ubiquitous computing, P2P computing and Ad hoc, etc., network is a dynamic and cooperative system made up of multi-software serving. Under the dynamic and uncertainty environment, traditional approaches to security are often found lacking, so dynamic trust model becomes a new and hot topic of security research for these new distributed applications. Focusing on dynamic, uncertainty and collaborative between entities under large-scale distributed environment, theory of Reinforcement Learning (RL) is applied to the study of dynamic trust model. First, basic formal description is conducted for trust decision, and behavior state-space structure is constructed based on gain function according the interaction time sequence between entities. Then, applying RL algorithm, based on direct trust gain function and feedback trust gain function, overall trust degree fusion computing model is set up. New model makes full use of the advantages of the RL algorithm, brakes away from the incongruence problem of trust-making in the traditional method, in which the weights are set up by subjective manners. Simulation's results show that, compared to the existing trust models, the new model has a better dynamic adaptation capability. © 2008 Binary Information Press.