Integrated atmosphere-to-seafloor monitoring needed to better forecast cyclones, researchers say

A death toll of more than 1,100 is expected to rise significantly after a rare combination of prolonged heavy rainfall and multiple tropical cyclones that devastated several countries in south and southeast Asia. Tropical cyclones — intense rotating storms with high winds over warm ocean waters — typically do not form within a few degrees of the Earth’s equator. During last week’s extreme weather, however, three storms developed unusually close to the equator in this region. As the frequency and intensity of powerful tropical cyclones increase in many ocean basins, a team of researchers is calling for better early warning systems through the development of intelligent observation networks to improve tracking and forecasting of such events.

They published a review of the current state, key features and operational progress of mobile observation platforms for tropical cyclones and their role in improving forecasts on Oct. 15 in Ocean-Land-Atmosphere Research.

“Tropical cyclones, particularly typhoons in the Northwest Pacific, represent one of the most destructive ocean–atmosphere interactions, posing severe threats to coastal infrastructure and human safety,” said corresponding author Han Zhang, professor at the Second Institute of Oceanography, Ministry of Natural Resources (MNR) in Hangzhou, China, and Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai) in Zhuhai, China. “Accurate forecasting requires high-resolution, real-time data, which traditional methods like manned aircraft and buoys struggle to supply due to high risks and sparse coverage. Recent advances in unmanned systems — including drones, uncrewed surface vehicles and underwater gliders — now enable integrated, multi-scale monitoring from atmosphere to seafloor.”

In their paper, the researchers summarize the capabilities and operational progress of observation platforms across four dimensions — atmosphere, sea surface, upper ocean and seafloor — with an emphasis on synergistic applications and data assimilation to improve forecasting. Current monitoring platforms in each of these areas have their own strengths and limitations, according to Zhang. For example, manned aircraft and research vessels face high operational costs and risks, while individual unmanned platforms cannot on their own provide continuous, three-dimensional coverage throughout a cyclone’s life cycle. As such, the researchers emphasized coordinated, multi-platform observation networks rather than relying on any single technology.

“A paradigm shift is underway in tropical cyclone observation,” Zhang said. “The limitations of conventional platforms have made it clear that reliable forecasting can no longer depend on singular data sources. Instead, the closure of critical data gaps demands an intelligent, multi-platform network that synergistically employs unmanned aerial, surface, and underwater vehicles. This integrated approach, creating a dynamic observation network across the atmosphere, ocean surface, and water column, represents the future of tropical cyclone science. The greatest forecasting gains will be realized through the coordinated deployment and data assimilation of this mobile unmanned fleet, rather than from any single technological advance.”

The researchers aggregated the platforms and devices used to monitor each observation dimension and the programs they use. They also examined which monitoring approach was used to observe aspects of different tropical cyclones. By assessing how they overlap, the researchers concluded that establishing robust communication links with minimal delays among uncrewed vehicles in the air, on the surface and underwater, observational data and platform status can be exchanged seamlessly, supporting coordinated multi-vehicle observations even within harsh tropical cyclone environments.

“These high-resolution, multi-platform observations can then be ingested directly into coupled atmosphere-ocean forecast models, providing more accurate initial conditions and significantly enhancing the prediction skill for tropical cyclone, track intensity and air-sea interaction processes,” Zhang said, explaining that the long-term vision is to drastically reduce tropical cyclone track and intensity forecast errors, enabling earlier and more reliable warnings for affected regions. “Achieving this requires observing air–sea interaction processes with unprecedented detail, capturing the fine-scale exchanges of heat, momentum and moisture that govern cyclone evolution. Ultimately, such advances directly support coastal hazard mitigation, the protection of critical infrastructure and long-term climate resilience.”

According to Zhang, this study provides a critical foundation for developing intelligent observation networks to enhance forecasting and coastal resilience.

“The goal is to build a multi-platform, cross-scale, and mobile ocean observation networks that ensure every tropical cyclone becomes a well-observed cyclone with no blind spots and no unknowns,” Zhang said.

Other contributors include first author Xia Rong, postgraduate, and Di Tian, associate professor, who are both with the Second Institute of Oceanography, MNR, and Southern Marine Science and Engineering Guangdong Laboratory.

This work was supported by the Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai), the Scientific Research Fund of the Second Institute of Oceanography, MNR, the Project of State Key Laboratory of Satellite Ocean Environment Dynamics, the Innovation Group Project of the Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai), the National Natural Science Foundation of China, the Key R&D Program of Zhejiang Province, the National Key R&D Program of China, the Key Laboratory of the Polar Atmosphere–Ocean–Ice System for Weather and Climate, Ministry of Education, and the Shanghai Frontiers Science Center of Atmosphere-Ocean Interaction, Fudan University.