Volume 59(2) – 2010

Improving cyclone warning
Case study: Philippines

by Paula McCaslin1, Tetsuo Nakazawa2, Richard Swinbank3 and Zoltan Toth1

line

Better cyclone prediction is a focus of international weather research. This case study of a 2009 typhoon that narrowly missed the Philippines gives a portrait of the scope for ensemble forecasting.

A key challenge for meteorologists of the 21st century is to improve prediction of severe weather events like Typhoon Parma, so that people can protect themselves through timely warnings.

When Typhoon Parma struck in September 2009, it was the worst storm in four decades in the Philippines, affecting more than 3  million people, killing 288 and causing over US$ 600 million of damage.

map  

Ensemble forecasting archives open to researchers.

Ten of the world’s leading weather forecast centres contribute ensemble forecasts to the THORPEX Interactive Grand Global Ensemble (TIGGE) project. The ten data providers transmit forecasts to three archive centres, available to scientific researchers worldwide. One project that uses this data is the Global Interactive Forecast System, which helps the international community to improve cyclone warnings.

© European Centre for Medium-Range Weather Forecasts    

Tropical cyclones, also known as hurricanes and typhoons, are the most powerful and destructive weather systems on the planet. While success achieved with numerical weather prediction is one of the most significant scientific achievements of the 20th century, there is room to improve forecasting for rare but severe events that cause catastrophic damage.

Building on the foundations of numerical weather prediction and ensemble forecasting, TIGGE is an example of the world’s leading weather forecast centres, collaborating together with the global meteorological community.

The aim is to develop Global Interactive Forecast System to provide the best possible forecasts for tropical cyclones and other high-impact weather events.

This project is part of a WMO international research programme, in place since 2003, called THORPEX, The Observing System Research and Predictability Experiment. The programme is part of the WMO World Weather Research Programme, and a key research component of the WMO Disaster Risk Reduction Programme.

To illustrate how the system can help with tropical cyclone forecasts, Typhoon Lupit makes a good case study. Just as people in the Philippines were responding to Typhoon Parma, Typhoon Lupit raged in the western Pacific Ocean in mid-October 2009, and threatened further devastation.

man in front of computer screens  

Regional distribution hubs.

New forecast products, based on data from several ensemble prediction systems, aim to produce the best possible forecasts of tropical cyclones. These are delivered to regional centres, which in turn deliver them to the national weather forecast services. They inform emergency services about strong winds, severe rain and storm surges associated with tropical cyclones, and together issue public warnings as required. Above, a forecaster at the regional centre in Tokyo.

Courtesy of Tetsuo Nakazawa, Meteorological Research Institute, Japan    

The satellite image of Typhoon Lupit of 17 October 2009 came at a time when TIGGE data providers had been running their ensemble prediction systems, and were starting to produce track forecasts (see image of ensemble forecasts for the typhoon). Most forecasts showed that the cyclone would track westwards to strike the Philippines, compounding recent devastation from Typhoon Parma. Over the next few days, Lupit did indeed track westwards, as shown by the black line.

Ensemble forecast information was also used to calculate strike probability maps, helping forecasters make decisions for issuing warnings. While most tracks showed that Lupit could strike the northern Philippines, some showed that the typhoon could turn northeast.

Later ensemble forecasts showed in­creasing probability that Lupit would turn northeast, sparing the Philippines this time – which is what actually happened.

The storm’s path and whether it will hit land are important to know. Equally important to determine are the amount of rain and how strong the destructive winds will be. Typhoons also often cause vast damage to coasts and islands, due to storm surges and flooding. Some of Lupit’s towering thunderstorms reached as high as 15 kilometres (more than 9 miles high), indicating very powerful storms with heavy rainfall. Future forecast products will highlight risks from intense rainfall and winds associated with severe tropical storms.

The case shows how ensemble forecasts can give estimates of likely forecast outcomes, as well as early indications of possible scenarios, to alert decision-makers to a range of scenarios for severe weather.

Quality is constantly improving, though some inaccuracy will always remain. The atmosphere is complex and there are theoretical and practical limits to predictability. Probability forecasts have uncertainty factored into them. This is the very reason that probabilistic forecasts are useful.

 

Would Typhoon Lupit strike the Philippines?

The question for the Philippines, two weeks after the devastating Typhoon Parma of 2009, was whether a second typhoon would strike the country’s northeastern corner.

By continuously generating and analysing ensemble forecasts, a global network of expert centres provided the best possible analysis as the second typhoon unfolded. In the first set of images, an ensemble forecast gave an early indication of two completely different possible tracks that Lupit could follow. Later information showed increasing probability that the Philippines would be spared.

globe  

Typhoon Lupit, 17 October 2009.

This Japanese satellite image, processed by the National Institute of Informatics of Japan, shows Typhoon Lupit moving towards the Philippines.

     
map  

Predicting the course of Typhoon Lupit.

The ensemble forecasts of Typhoon Lupit’s cyclone track come from several TIGGE data providers. The colour changes every 24 hours along each forecast track.

     
map   Early ensemble forecasts indicated two completely different tracks that Lupit could follow. This later version showed that it was more likely that the typhoon would veer northeast, sparing the Philippines. This picture shows a NOAA Website display of hurricane forecasts tracks from three TIGGE data providers for Lupit (in colour, with actual track in black).

 

 

In other words

Numerical weather prediction

Modern weather forecasts are based on the technique known as numerical weather prediction:

  • First, observations are collected to determine weather conditions;
  • Next, advanced high-speed computers solve complex sets of numerical equations, using the weather observations as variables;
  • This representative model of the atmosphere is used to predict the future state of the atmosphere at specific time intervals.

Ensemble Forecasting

Ensemble probabilistic forecasting is an advanced application of numerical weather prediction. This powerful new tool can improve early warning of high-impact events:

  • Typically, 20 or more numerical model predictions are generated for a given time;
  • Each individual prediction is known as an ensemble member;
  • The ensemble of multiple predictions captures the range of possible weather events and impact scenarios;
  • Decision-makers use the information to issue warnings with longer lead times, reducing losses and increasing safety.

Ensemble Prediction Systems are now common at most of the major operational weather prediction facilities worldwide.

 

 

1 Global Systems Division, Earth Systems Research Laboratory, National Oceanic and Atmospheric Administration, USA

2 Typhoon Research Department, Meteorological Research Institute, Japan

3 Ensemble Forecasting Group, UK Met Office

back to top

 

 

 

 

 

 


Download
   
  Latest issue [pdf]




Copyright | Privacy policy | Disclaimer | Guidelines |

 

.