Fig 2 - uploaded by Tangao Hu
Content may be subject to copyright.
Source publication
The location of a typhoon center plays an important role in moving track monitoring and forecasting. More and more new satellites are used to monitor typhoons. Combining multiple sources of satellite data has become popular for tropical cyclone monitoring in recent years. In this letter, we demonstrate a robust method of locating the typhoon center...
Similar publications
Conventional scatterometer wind retrieval has been proved effective for low and moderate wind conditions. Under high wind conditions, especially for typhoons, the air–sea interactions are fierce, and usually the sea surface backscattering is complicated by rain, leading to degraded scatterometer wind quality. Under such circumstances, radiometers i...
Citations
... In addition, TC intensity estimation, TC tracking, and TC forecasting require accurate TC center location (Olander andVelden 2007, 2019). Therefore, accurate TC center location is crucial for forecasters and emergency responders (Jaiswal and Kishtawal 2013;Hu et al. 2017;Lu et al. 2017). ...
In this paper, a data-driven transfer learning (TL) model for locating tropical cyclone (TC) centers from satellite infrared images in the Northwest Pacific is developed. A total of 2450 satellite infrared TC images derived from 97 TC between 2015 and 2018 were used for this paper. The TC center location model (ResNet-TCL) with added residual fully-connected modules is built for the TC center location. The MAE of the ResNet-TCL model is 34.8 km. Then TL is used to improve the model performance, including obtaining a pre-trained model based on the ImageNet dataset, transferring the pre-trained model parameters to the ResNet-TCL model, and using TC satellite infrared imagery to fine-train the ResNet-TCL model. The results show that the TL-based model improves the location accuracy by 14.1% (29.3 km) over the No-TL model. The model performance increases logarithmically with the amount of training data. When the training data are large, the benefit of increasing the training samples is smaller than the benefit of using TL. The comparison of model results with the Best Track data of TCs shows that the MAEs of TCs center is 29.3 km for all samples and less than 20 km for H2-H5 TCs. In addition, the visualization of the TL-based TC center location model shows that the TL model can accurately extract the most important features related to TC center location, including TC eye, TC texture, and contour. On the other hand, the No-TL model does not accurately extract these features.
... Related works based on fluxes of the gradient vectors of brightness temperature and fitting spiral features within the IR images are also used for fixing the center position of TCs [14], [15]. Hu et al. [16] presented as an approach for locating the TC center by using satellite and microwave scatterometer data. Furthermore, machine learning (ML)-based various frameworks for land use-land cover classification from remote sensing are presented in [17]. ...
Tropical cyclones (TCs) are the most destructive weather systems that form over the tropical oceans, with about 90 storms forming globally every year. The timely detection and tracking of TCs are important for advanced warning to the affected regions. As these storms form over the open oceans far from the continents, remote sensing plays a crucial role in detecting them. Here we present an automated TC detection from satellite images based on a novel deep learning technique. In this study, we propose a multi-staged deep learning framework for the detection of TCs, including, (i) a detector - Mask Region-Convolutional Neural Network (R-CNN), (ii) a wind speed filter, and (iii) a classifier - CNN. The hyperparameters of the entire pipeline is optimized to showcase the best performance using Bayesian optimization. Results indicate that the proposed approach yields high precision (97.10%), specificity (97.59%), and accuracy (86.55%) for test images.
... Related works based on fluxes of the gradient vectors of brightness temperature and fitting spiral features within the IR images are also used for to fix the center position of TCs [11], [12]. [13] presented as an approach for locating the typhoon center by using satellite and microwave scatterometer data. ...
div>
Tropical cyclones (TCs) are the most destructive weather systems that form over the tropical oceans, with 90 storms forming globally every year. The timely detection and tracking of TCs are important for advanced warning to the affected regions. As these storms form over the open oceans far from the continents, remote sensing plays a crucial role in detecting them. Here we present an automated TC detection from satellite images based on a novel deep learning technique. In this study, we propose a multi-staged deep learning framework for the detection of TCs, including, (i) a detector - Mask Region-Convolutional Neural Network (R-CNN), (ii) a wind speed filter, and (iii) a classifier - CNN. The hyperparameters of the entire pipeline is optimized to showcase the best performance using Bayesian optimization. Results indicate that the proposed approach yields high precision (97.10%), specificity (97.59%), and accuracy (86.55%) for test images.
</div
... Related works based on fluxes of the gradient vectors of brightness temperature and fitting spiral features within the IR images are also used for to fix the center position of TCs [11], [12]. [13] presented as an approach for locating the typhoon center by using satellite and microwave scatterometer data. ...
div>
Tropical cyclones (TCs) are the most destructive weather systems that form over the tropical oceans, with 90 storms forming globally every year. The timely detection and tracking of TCs are important for advanced warning to the affected regions. As these storms form over the open oceans far from the continents, remote sensing plays a crucial role in detecting them. Here we present an automated TC detection from satellite images based on a novel deep learning technique. In this study, we propose a multi-staged deep learning framework for the detection of TCs, including, (i) a detector - Mask Region-Convolutional Neural Network (R-CNN), (ii) a wind speed filter, and (iii) a classifier - CNN. The hyperparameters of the entire pipeline is optimized to showcase the best performance using Bayesian optimization. Results indicate that the proposed approach yields high precision (97.10%), specificity (97.59%), and accuracy (86.55%) for test images.
</div
... Satellite data are obtained from the National Satellite Meteorological Center (China). The FY-2 series is the first generation of Chinese geostationary meteorological satellites, which has a frequency of half an hour for producing the cloud images [32]. In this study, we used the satellite cloud images right on every hour. ...
... To compensate for the shortcoming of the uneven spatial distribution of the AWSs, this study also uses reanalysis data from the European Centre for Medium-Range Satellite data are obtained from the National Satellite Meteorological Center (China). The FY-2 series is the first generation of Chinese geostationary meteorological satellites, which has a frequency of half an hour for producing the cloud images [32]. In this study, we used the satellite cloud images right on every hour. ...
Based on ten-year tropical cyclones (TCs) observations from 2009 to 2018, the black body temperature (TBB, also called cloud-top brightness temperature) data obtained from the infrared channel 1 (with the wavelength of 10.30–11.30 µm) of the FY-2 satellite image, and the wind observation data at the automatic weather stations (AWSs) in Guangdong province, this study explores the relationship between the TBBs and the winds induced by TCs at AWSs. It is found that the wind speeds at AWSs cannot be obtained directly by using TBB value inversion, but the maximum potential wind gust (MPG) and the maximum potential average-wind (MPAW) at AWSs can be estimated when a TBB is known. Influenced by the terrain, the surrounding environment, and detected height, the MPG and the MPAW values of different AWSs may differ for the same TBB. The wind data from ERA5 reanalysis is also used to explore the relationship between the TBBs and the winds over grids area during the TCs’ periods. Similar to the AWSs, there is a capping function between the winds over the grids and the TBBs. The reanalysis data can generally show the average wind conditions of the weather stations inside the grids, and therefore, can be used to supplement the data for the areas where there is no AWS observation available. Such a study could provide references for estimating the potential wind disasters induced by TCs in the study area.
... Geng [8] proposed early warning method of overhead transmission line to predict the damage probability of overhead transmission line based on geographic information and meteorological information. Typhoon data can be monitored through meteorological satellite [9], radar [10], microwave radiometric observation [11] et al. These observation methods are designed for meteorological researches, and focused on the macro characteristics of typhoon. ...
Extreme weather like typhoon leaves significantly terrible influence on the stability of power grid in the coastal area. A high-voltage transmission line includes hundreds of towers and usually stretches hundreds of kilometers. It is not economical to install typhoon monitoring system throughout the transmission line. This paper proposes a typhoon inversion method (TIM) for transmission line which utilizes engineering wind field model and only requires monitoring data from a few towers. The developed method is called YanMeng wind field (YM) with directional mutation genetic algorithm (DMGA) for transmission lines, or YM-DMGA method. DMGA utilizes real-time monitoring data, and improves simulation accuracy of average wind speed by dynamically optimizing two critical parameters B and z0 in wind field model, which is a great improvement over the traditional YM model relying on historical data. The TIM is mainly composed of three parts, the meteorological monitoring system (MMS), the YMDMGA model, and the software system. The TIM collects real-time measured wind data of particular monitoring stations through the MMS. Meanwhile, wind data of the whole transmission line is simulated using the YM-DMGA and displayed by the software system. Then the method proposed is verified to be effective in three aspects. First of all, the YM-DMGA method has great enhancement in accuracy, for the coefficient of determination Rsquare increases from 0.811 to 0.986. Secondly, DMGA in the method has quicker convergent speed than typical GA, with fitness converged in the 269th epoch, earlier than that for typical GA, 656th epoch. Lastly, DMGA has better optimal fitness value of 0.6239, bigger than that of typical GA, 0.5891. At the end of the paper, an application of the method to an 110kV double-circuit transmission line is presented to reveal the risk under the impact of the super typhoon Rammasun.
... Among them, Tian et al. [7] design an adaptive threshold processing method to segment typhoon cloud system by using two value image connecting region labeling algorithm. Hu et al. [8] demonstrate a robust method of locating the typhoon center based on meteorological satellite and microwave scatterometer data. Xu et al. [9] propose an automatic method to determine the center of tropic cyclones from a serious of SAR images. ...
The task of typhoon center location plays an important role in typhoon intensity analysis and typhoon path prediction. Conventional typhoon center location algorithms mostly rely on digital image processing and mathematical morphology operation, which achieve limited performance. In this paper, we proposed an efficient fully convolutional end-to-end deep neural network named TCLNet to automatically locate the typhoon center position. We design the network structure carefully so that our TCLNet can achieve remarkable performance base on its lightweight architecture. In addition, we also present a brand new large-scale typhoon center location dataset (TCLD) so that the TCLNet can be trained in a supervised manner. Furthermore, we propose to use a novel TCL+ piecewise loss function to further improve the performance of TCLNet. Extensive experimental results and comparison demonstrate the performance of our model, and our TCLNet achieve a 14.4% increase in accuracy on the basis of a 92.7% reduction in parameters compared with SOTA deep learning based typhoon center location methods.
... There are, on average, 26-28 tropical storms and typhoons each year, two or three of which can reach typhoon intensity. Typhoons here are seasonal and usually occur from July to September [14]. We chose three representative typhoons (Usagi, Fitow, and Nari) that occurred in 2013 to validate the effectiveness of the proposed model. ...
Timely and accurate sea surface wind field (SSWF) information plays an important role in marine environmental monitoring, weather forecasting, and other atmospheric science studies. In this study, a piecewise linear model is proposed to retrieve SSWF information based on the combination of two different satellite sensors (a microwave scatterometer and an infrared scanning radiometer). First, the time series wind speed dataset, extracted from the HY-2A satellite, and the brightness temperature dataset, extracted from the FY-2E satellite, were matched. The piecewise linear regression model with the highest R2 was then selected as the best model to retrieve SSWF information. Finally, experiments were conducted with the Usagi, Fitow, and Nari typhoons in 2013 to evaluate accuracy. The results show that: (1) the piecewise linear model is successfully established for all typhoons with high R2 (greater than 0.61); (2) for all three cases, the root mean square error () and mean bias error (MBE) are smaller than 2.2 m/s and 1.82 m/s, which indicates that it is suitable and reliable for SSWF information retrieval; and (3) it solves the problem of the low temporal resolution of HY-2A data (12 h), and inherits the high temporal resolution of the FY-2E data (0.5 h). It can provide reliable and high temporal SSWF products.
... T. Hu et. al. [45] have used the Feng Yun-2 (FY-2) satellite and microwave scatterometer Hai Yang-2 (HY-2) to detect the center of a typhoon successfully. The northwest region of the Pacific Ocean is chosen for study in this article due to a high rate of cyclones and typhoons hitting the region every year. ...
... Therefore they have stated the SST is not a justifiable parameter to monitor cyclones in the Indian Ocean. [45] 18 km to 100 km S and X bands Wind Speed, SST, Ice monitoring and Water vapour monitoring Neeru et. al. [50] have experimented on the images of a couple of cyclones that hit India namely the Phyan, Ward, Laila and Phet. ...
... At present, the observation of extreme events such as tropical cyclone, tropical storm and tropical depression through satellite measurements is mainly performed through visualization of large cyclonic cloudy and infrared temperature data (X Li et al., 2013;Hu et al., 2017). The small number of tide gauges and unavailable buoy data are surely cannot represent the large region of the South China Sea and restrict the analysis of the tropical cyclone signature in this area. ...
... For a global perspective, Chen et al. (2002) were used collocated satellites altimeter and scatterometer to observe the spatial and seasonal pattern of dominant swell and wind wave. Besides, there are other parameters from different satellite sensors can be exploited to deduce the development of the tropical cyclone (Hu et al., 2017). ...
Continous and synoptic mapping of sea level anomaly (SLA) across the marginal seas is possible thanks to the abundance of altimeter satellite with regulated orbits around the Earth. Though misestimation of satellite derived SLA due to spatial and temporal variation is evident and has led to the significant total bias on the final estimation. The SLA estimation in the South China Sea (SCS) helps in deducing the signature impact of tropical cyclones (TC) in which the TC derived anomalous sea level height can be distinguished. Sparse and small number of tide gauges and buoys available in the SCS limit the spatial and temporal distribution for sea-truth SLA acquisition. This limitation becomes more significant when the single mission satellite data produces SLA data with non-reliable and void pixels. Therefore, this paper is designed to preliminarily review the potentials and challenges of applying satellite altimeter derived SLA from single and multi-mission to deduce the tropical cyclone signature in the SCS. For this study, three altimeter satellite missions namely Jason-1, Envisat-1 and Geosat Follow-On were incorporated to acquire SLA during the severe and energetic TC event Typhoon Chanchu 2006. The results show good agreement of interpolated SLA maps with the TC event nearby at the time of occurrence of SLA variation. This study concludes that multi-mission satellite data has a prominent ability for synoptic mapping the sea level anomaly variation during the tropical cyclone activities in South China Sea.
