Atmospheric Sciences Project Update

New Data Buoys Watch Typhoons from Within the Storm

Advanced real-time data buoys have observed nine strong typhoons in the northwestern Pacific Ocean since 2015, providing high-resolution data and reducing the uncertainty of numerical model forecasts.

By Sen Jan, Yiing Jang Yang, Hung-I Chang, Ming-Huei Chang, and

Two recently deployed ocean data buoys got a real test of their abilities last July when category 5 “super typhoon” Nepartak’s center passed within a few kilometers of where they were moored. This meeting was no accident: the buoys were specifically designed to record data on passing typhoons and relay the information to Taiwan’s Central Weather Bureau.

Typhoons, hurricanes that occur in the northwestern Pacific Ocean, threaten coastal and inland populations of nearly 1 billion people living in Taiwan, Japan, Korea, China, and the Philippines. Broadcasting timely warning messages to the public is an essential part of minimizing the disastrous effects of typhoons in coastal and inland areas. However, accurate and timely warnings require better forecasting of typhoon tracks and intensities. High-resolution data and real-time observations of typhoons in the open ocean are important to better understand their formation and evolutionary processes and to improve the performance of numerical predictions.

Although satellite remote sensing is valuable, these sensors cannot see far beneath the sea surface, making a real-time data buoy network with subsurface observations the perfect complement during typhoons. However, establishing a high density of buoy stations requires a large number of reliable instruments, as well as considerable support in the form of mooring supplies, satellite communications, and ship time.

A research and technical team composed of scientists and technicians at National Taiwan University (NTU) is working to establish just such a buoy network in the northwest Pacific. The team is working to reduce the cost of data buoys and their associated maintenance using low–power consumption technologies, building on the basic concept of the Autonomous Temperature Line Acquisition System (ATLAS) [Hayes et al., 1991]. The new data acquisition system accommodates more meteorological and oceanic sensors than a standard ATLAS mooring and can transmit data at adjustable intervals via satellite-to-ground stations.

Economical, Reliable Observation and Forecasting

Typhoons are capable of causing catastrophic destruction along their path. Typhoon Morakot, a category 2 storm, passed across southern Taiwan on 8 August 2009, bringing the highest typhoon-related rainfall to Taiwan in 50 years. The heavy rain caused floods and mudslides, resulting in more than 600 deaths and US$5 billion in damages [National Disasters Prevention and Protection Commission, 2009]. On 8 November 2013, Super Typhoon Haiyan ripped through the central Philippine islands, causing more than 6000 deaths and US$2 billion in losses [Shull, 2013].

The threat of such death and destruction creates an obvious need for buoy observations in the Philippine Sea. Unfortunately, these kinds of on-site observations are rare, presumably because of the high cost of setting up and maintaining deepwater buoy stations, the frequent failure of the buoys in extreme conditions, and the instrument damage and buoy relocation that can happen when the buoys become entangled in tuna fishing lines.

The buoy project, initiated by the Institute of Oceanography at NTU, aims to overcome these difficulties with a new, less expensive but more robust data buoy, which we call the NTU buoy. The new data collected by these buoys will help improve our understanding of the ways in which the air and sea exchange heat and how this exchange influences typhoon intensity. The NTU buoys will also provide real-time data to weather forecast centers when a typhoon is approaching.

A New Buoy System Design

The NTU buoy components, with modules for surface and subsurface data acquisition and for satellite and radio communications.
Fig. 1. The NTU buoy components, including the associated solid-state modules for surface and subsurface data acquisition and for satellite and radio communications.

The NTU buoy has a newly designed data acquisition system, electric power scheme, and satellite communications module (Figure 1), making it an improvement over the standard ATLAS mooring. The data acquisition system is controlled by a microprocessor that features low electric power consumption. The meteorological sensors can be in sleep mode between successive samplings to save energy.

Lithium batteries supply the buoy with electric power, and they can support buoy operations for more than 18 months. Solar panels, with their stainless-steel towers, are not required for this buoy system, so the wind drag on the buoy is greatly reduced.

Previous buoy systems have become entangled in fishing lines, but the new design minimizes this problem with unique cutter devices clamped onto the mooring cable. These cutters have four stationary blades that radiate out from the clamp, and they sever any fishing line that drapes across them before the line can become entangled around the mooring cable or the buoy.  A customized stainless-steel connector and a high-density polyethylene-coated steel wire in the upper 500 meters of the approximately 5500-meter mooring line help the buoy survive in strong typhoons.

Two separate data loggers record air temperature, pressure, wind speed and direction, humidity, solar radiation, irradiance, precipitation, and the seawater temperature and salinity profile in the upper 500 meters of the ocean every 6 minutes. The system also measures ocean current velocity at 25- and 75-meter depths. The data can be transmitted to the home station every 30 minutes during normal weather conditions and can be remotely adjusted to transmit every 12 minutes during the passage of a typhoon via satellite communication. Stand-alone time-lapse cameras (Figure 1) are mounted on the buoy to take sea surface images at 1-minute intervals.

Scientific Goals

Understanding typhoon dynamics is still one of the key challenges that limits the performance of forecast models. Our buoy project provides high-resolution measurements of the oceanic and atmospheric responses to typhoons, and it enables comparison of these observations with the forecasts produced by numerical models.

We hope to answer the following science questions [D’Asaro et al., 2013] as we move forward with this buoy program:

  • What is the buoy’s best wind estimate in a typhoon?
  • How does the ocean’s mixed layer, where turbulence mixes the water, evolve under extreme typhoon winds?
  • How do extremely strong winds affect the exchange of heat, salt, and momentum between the sea and the air?
  • How can these new buoy data improve forecasting?

Into the Path of the Storm

To answer these questions, the NTU team took the most likely typhoon track into consideration. The team deployed a prototype NTU buoy 375 kilometers off southeastern Taiwan (123.9°E, 21.2°N) during June–September 2015. In June–October 2016, we deployed two improved buoys, NTU1 and NTU2, 375 kilometers (123.9°E, 21.1°N) and 175 kilometers (122.6°E, 21.9°N), respectively, off the coast of southeastern Taiwan.

These buoys measured nine typhoons. The first four, Chan-hom, Linfa, Soudelor, and Goni, occurred in 2015, and the later five, Nepartak, Meranti, Malakas, Megi, and Aere, occurred in 2016.

Data collected by the buoys were sent in real time to Taiwan’s Central Weather Bureau. These data provided valuable information on air-sea fluxes for use in improving forecast model output for validating and calibrating satellite observations during these typhoons.

Data from a Super Typhoon

Notably, two NTU buoys observed Typhoon Nepartak from their respective locations, which proved to be only a few kilometers from the typhoon’s center (Figure 2). This typhoon became a category 5 super typhoon on 6 July 2016, and it reached NTU1 with a well-formed, distinguishable eye. As the eye of this storm approached NTU1, the buoy observed an atmospheric pressure of 940 hectopascals (standard sea level air pressure is 1013 hectopascals), a maximum wind gust of 41 meters per second (about 148 kilometers per hour), and a decrease of seawater temperature from 31°C to 28°C in the upper 100 meters. The typhoon’s eye soon reached NTU2, and this second buoy recorded a rather low atmospheric pressure of 911 hectopascals and a maximum wind gust of 44 meters per second (about 158 kilometers per hour). The seawater in the upper 120 meters was well mixed by the strong winds (see temperature profile in Figure 2).

The track of Typhoon Nepartak and the locations of the two data buoys, NTU1 and NTU2.
Fig. 2. This image of Typhoon Nepartak was captured by the satellite Himawari-8 on 7 July 2016 at 02:30 and 12:00 coordinated universal time (UTC). The typhoon track is shown, along with the locations of the two buoys, NTU1 and NTU2. The insets show buoy-measured wind velocities (red lines) and seawater temperatures (colored depth profiles) in the upper 250 meters for NTU1 (inset a) from 21:00 UTC on 6 July 2016 to 09:00 UTC on 7 July 2016 and from NTU2 (inset b) from 06:00 UTC on 7 July 2016 to 18:00 UTC on 7 July 2016.

Next Steps

We successfully recovered the NTU buoys on 14 October 2016. We are currently performing quality assurance and quality control operations on the complete data set that we retrieved from the buoys at that time in preparation for making these data available to the broader scientific community.

We are also in a stage of developing and testing prototypes, and the buoy observation serves as a modest start toward a more widespread network. In summer 2017, we plan to deploy two more NTU buoys. Our ultimate goal is to complete and maintain five operational buoys in this region in the next 3 years.


The buoy project is sponsored by the NTU under the Aim for the Top University Plan. Taiwan’s Ministry of Science and Technology and Central Weather Bureau partially support this project. The technicians at NTU and the crew of R/V Ocean Researcher I helped deploy, maintain, and recover the buoys. H.-I.C. is supported by MOST 104-2611-M-002-012-MY2. We dedicate this report to Wen-Huei Lee, who was one of the technicians of the NTU buoy team.


D’Asaro, E. A., et al. (2013), Impact of typhoons on the ocean in the Pacific, Bull. Am. Meteorol. Soc., 95(9), 1405–1418,

Hayes, S. P., et al. (1991), TOGA TAO: A moored array for real-time measurements in the tropical Pacific Ocean, Bull. Am. Meteorol. Soc., 72, 339–347,<0339:TTAMAF>2.0.CO;2.

National Disasters Prevention and Protection Commission (2009), Typhoon Morakot disaster responses reports from Typhoon Morakot Central Emergency Operating Center, 74th report, Taipei, Taiwan, Morakot Disaster Responses Reports_74_20090908_1830.pdf.

Shull, P. (2013), Typhoon Haiyan damage summary, Global Agricultural Information Network report, Foreign Agric. Serv., U.S. Dep. of Agric., Washington, D. C.

Author Information

Sen Jan, Yiing Jang Yang, Hung-I Chang, Ming-Huei Chang, and Ching-Ling Wei (email: [email protected]), Institute of Oceanography, National Taiwan University, Taipei, Taiwan

Citation: Jan, S., Y. J. Yang, H.-I. Chang, M.-H. Chang, and C.-L. Wei (2017), New data buoys watch typhoons from within the storm, Eos, 98, Published on 27 March 2017.
© 2017. The authors. CC BY-NC-ND 3.0