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.

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.

As a category-3 typhoon, Hato (2017) experienced the notable rapid intensification (RI) over the hot sea surface before its landfall. The RI process and the influences of local sea surface temperature (SST) patterns on the evolution of Hato were well captured and carefully investigated using a high-resolution air–sea coupled model. To further explore the close relationship between the radial distributions of SST and storm evolution, a sensitive experiment with time-fixed SST was also performed. Results showed that the time-fixed SST experiment produced earlier RI following the rapid core structure adjustment, as higher SST in the core region was found favorable to increasing the near-surface water vapor and latent heat flux. Strong updrafts were thus facilitated inside the eyewall, inducing the eyewall contraction and RI of the storm. In contrast, cooler SST inside the core region should account for the delay of RI as the intense convection located in the outer rainbands, inhibiting the transportation of energy into the inner-core. Momentum tendency analysis also proves these mechanisms. Therefore, not only the value of SST but also its radial-gradient, plays an important role in the evolution of tropical cyclones, highlighting the need for an advanced air–sea coupled model.

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.

Multi-satellite and buoy observation data were used to systematically analyze the ocean response offshore of Taiwan to Super Typhoon Nepartak in 2016. The satellite data showed that a high sea surface temperature combined with a thick warm water layer and deep mixed layer provided a good thermal environment for continuous intensification of the typhoon. Two high-resolution buoys (NTU1 and NTU2) moored 375 and 175 km offshore of southeastern Taiwan were used to clarify the typhoon–ocean interaction as the typhoon approached Taiwan. The ocean conditions were similar at the two buoys before the typhoon, and both buoys were on the left side of the typhoon track and suffered similar typhoon factors (e.g., typhoon intensity and translation speed) during its passage. However, the ocean response differed significantly at the two buoys. During the forced period, the entire upper ocean was cooled at NTU1. In contrast, there was a clear three-layer vertical structure at NTU2 consisting of cool surface and deep layers with a warmer layer between the two cool layers. These responses can be attributed to strong upwelling of a cold eddy at NTU1 and vertical mixing at NTU2. These results indicate that, under similar preexisting conditions and typhoon factors, the movement of ocean eddies under typhoon forcing is an unexpected mechanism that results in upwelling and thus needs to be considered when predicting changes in the ocean environment and typhoon intensity.