Relationship between sea surface temperature and the maximum intensity of tropical cyclones affecting Vietnam's coastline

Thanh, Nguyen Thi; Cuong, Hoang Duc; Hiền, Nguyễn Xuân; Kieu, Chanh

doi:10.1002/joc.6348

Cited by 7 publications

(4 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We will further add temporal dimension information to the dataset by combining CNN with the long short term memory 31 model to effectively extract temporal features from adjacent TC satellite images to further improve the accuracy of TC intensity estimation. The present intensity estimation model does not take into account physical factors, such as vertical wind shear 32 and ocean surface temperature, 33 which may modulate rapid TC intensity changes. In the future, the above physical features will be added to our model to improve its TC intensity estimation performance.…”

Section: Discussionmentioning

confidence: 99%

Estimating tropical cyclone intensity using dynamic balance convolutional neural network from satellite imagery

Tian

Lai

Niu³

et al. 2023

J. Appl. Rem. Sens.

View full text Add to dashboard Cite

Accurate estimation of tropical cyclone (TC) intensity helps to understand the evolution of TCs throughout their life cycle and plays an essential role in mitigating TC impact. Although TC intensity estimation methods based on deep learning have made significant progress, the existing techniques do not apply good methods to overcome the intensity overestimation and underestimation problems caused by the unbalanced distribution of TC data. Therefore, we propose a dynamic balance convolutional neural network to overcome these issues. The model consists of two branches, one branch is the learning of the raw data, and the other is the learning of strong (weak) TCs that account for a few data samples. Finally, the model is dynamically adjusted by adaptive trade-off parameters, gradually from the learning of the raw data to the learning of strong (weak) TCs, thus reducing errors in underestimation (overestimation) of strong (weak) TCs. Furthermore, an attention mechanism is employed to obtain the correlation between channels to improve the accuracy of TC intensity estimation further. We used globally 1285 TC cases from 2003-2016 to train the model and globally 94 TC cases from 2017 as independent test data. The results showed that the root-mean-square error of TC intensity estimation was 8.32 kt, 35% lower than that of the advanced Dvorak technique and 26% lower than that of the deep learning method visual geometry group (VGG). For a subset of 482 samples (from East Pacific and Atlantic) analyzed with reconnaissance observations, a root-mean-square intensity difference of 7.95 kt is achieved. Finally, we explored the model's feature learning process and the contribution of each component of the satellite image to the TC intensity estimation through model visualization.

show abstract

Section: Discussionmentioning

confidence: 99%

Estimating tropical cyclone intensity using dynamic balance convolutional neural network from satellite imagery

Tian

Lai

Niu³

et al. 2023

J. Appl. Rem. Sens.

View full text Add to dashboard Cite

show abstract

“…In addition to challenges at the local level, Phu Quoc MPA faces challenges at a global scale (Pratchett et al, 2008; Smale et al, 2019). Vietnam is highly vulnerable to the impacts of climate change, which directly threatens the livelihoods of coastal communities and the ecosystems upon which they depend (Intergovernmental Panel on Climate Change, 2014; Thanh et al, 2020; Huynh et al, 2021). Rising sea levels, the inundation of floodwaters, beach erosion, and marine heatwaves are expected to intensify under climate change (Pratchett et al, 2008; Minderhoud et al, 2019; Smale et al, 2019; The World Bank, 2021).…”

Section: Discussionmentioning

confidence: 99%

Marine protected areas ineffectively protect seagrass and coral reef fish communities in the Phu Quoc and An Thoi archipelago, Vietnam

Tran¹,

Carve

Dong³

et al. 2022

Aquatic Conservation

View full text Add to dashboard Cite

1. The Phu Quoc marine protected area (MPA) is a multiple-use MPA that was established in 2007 and is a component of Vietnam's National MPA System. The MPA is divided into two spatially separated zones based on habitat type: a seagrass zone and a coral reef zone.2. In this study, visual census data were collected in 2018-2019 and were used to derive fish biomass and community diversity metrics as proxies for ecosystem condition and function (trophic and mobility). The effectiveness of the MPA in protecting fish communities in the seagrass and coral reef zones was evaluated.An updated characterization of the Phu Quoc fish communities, and an assessment of fish community variation among regions and between habitat types, is provided.3. In total, 125 species representing 74 genera of 40 families were recorded.Omnivores were the most abundant (38%), followed by planktivores (24%), whereas piscivores (2%) were scarce. There was an especially low density of high-value fishery target species such as Haemulidae (<0.01%), Carangidae (<0.2%), Lethrinidae (<0.3%), Lutjanidae (<0.4%), Serranidae (<2%), and Chaetodontidae (<3%). Alarmingly, fishes with a total length of >20 cm only accounted for 1% of the total individuals. Species and functional composition were similar between protected and unprotected areas but differed among regions and between habitats.4. These results suggest that the protection of fishes provided by Phu Quoc MPA at the time of this study was ineffective. The taxonomic composition of fish communities has remained relatively unchanged over the last approximately 25 years, with the density of large fishes and commercial species remaining low.There is capacity to improve MPA performance by providing adequate resources for threat management and using evidence-based decision-making in management. The 2020 rezoning and expansion of the Phu Quoc MPA is discussed with regards to prioritizing biodiversity and coverage targets.

show abstract

“…In the future, we will consider combining other data, such as water vapor channel data and microwave data, and make full use of the advantages of various data to improve the performance of the model. 4) Currently, our intensity estimation model does not consider some physical factors, such as vertical wind shear [51], [52], ocean surface temperature [53], latitude and longitude of the TC center [36], and TC size [54], even though they may be important influencing factors for the rapid change of TC intensity. In the future, the above physical characteristics will be added to our model in order to improve its TC intensity estimation performance, and especially to provide new ideas for investigating sudden changes of TC intensity.…”

Section: Discussionmentioning

confidence: 99%

Tropical Cyclone Intensity Classification and Estimation Using Infrared Satellite Images With Deep Learning

Zhang

Wang

et al. 2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

A novel tropical cyclone (TC) intensity classification and estimation model (TCICENet) is proposed using infrared geostationary satellite images from the northwest Pacific Ocean basin in combination with a cascading deep convolutional neural network (CNN). The proposed model consists of two CNN network modules: a TC intensity classification (TCIC) module and a TC intensity estimation (TCIE) module. First, the TCIC module is utilized to divide TC intensity into three categories using infrared satellite images. Next, three TCIE models based on the CNN regression network that combine different intensity types of infrared satellite images with the TC best track data are presented. The three TCIE models consider classification error with the TCIC module in order to improve TCIE accuracy. A total of 1001 TCs from 1981-2019 were used to verify the proposed TCICENet model, with 844 TCs from 1981-2013 employed as training samples, 76 TCs from 2014-2016 used as validation samples, and 81 TCs from 2017-2019 used as testing samples. In order to reduce the computation burden of training the TCICENet model, various input image sizes were explored. An image size of 170 × 170 pixels achieved the best performance, with an overall root mean square error of 8.60 kt and a mean absolute error of 6.67 kt compared to the best track. Index Terms-Deep convolutional neural network (CNN), intensity estimation, intensity grade classification, tropical cyclone (TC). I. INTRODUCTIONT HE northwest Pacific Ocean basin is one of the most active tropical cyclone (TC) areas in the world, generating approximately 27 TCs per year. According to the TC standards (http://agora.ex.nii.ac.jp/digital-typhoon/help/unit.html. ja#id2) released by the Japan Meteorological Agency (JMA), TCs can be divided into six grades: tropical depression (TD; ∼33 kt), tropical storm (TS; 34-47 kt), severe tropical storm

show abstract

Relationship between sea surface temperature and the maximum intensity of tropical cyclones affecting Vietnam's coastline

Cited by 7 publications

References 42 publications

Estimating tropical cyclone intensity using dynamic balance convolutional neural network from satellite imagery

Estimating tropical cyclone intensity using dynamic balance convolutional neural network from satellite imagery

Marine protected areas ineffectively protect seagrass and coral reef fish communities in the Phu Quoc and An Thoi archipelago, Vietnam

Tropical Cyclone Intensity Classification and Estimation Using Infrared Satellite Images With Deep Learning

Contact Info

Product

Resources

About