Oct 12, 2006
Wind shifts make predictions hard: NEA
Weather forecast is tricky in the tropics where patterns are weak
By Arti Mulchand
WHY is the crystal ball so hazy, one might ask. It would be better if the weatherman could say in advance if haze levels were going to go up or down.
But the fact is that the light and constantly changing winds, characteristic at this time of the year, make it hard to predict what will happen, said the National Environment Agency (NEA). The region is experiencing the tail end of the south-west monsoon, and will soon enter the 'inter-monsoon' season, so the winds are weak - five knots or less, compared to the usual 10 to 15 knots - and variable.
'Even a change in 20 degrees in the wind direction could make a huge difference in terms of which part of South-east Asia would be affected,'' said Mr Lam Keng Gaik, NEA's chief meteorologist.
The winds are expected to continue shifting until later this month. That could also explain why some of NEA's predictions have not always been spot-on. On Friday, for instance, NEA had said the Pollution Standards Index (PSI) was unlikely to go beyond 100. The next day, it hit 150.
The wind is one of two key factors that determine how clear the skies will be. The other is the hot spots at the source - south Sumatra, Jambi and the Riau Islands. In any case, weather predictions are 'never 100 per cent' accurate, Mr Lam said.
On average, weather conditions in the tropics can be 'meaningfully predicted' - which means with 70 to 80 per cent accuracy - for one to three days. Forecasting is usually based on a combination of observation - looking at the actual burning and weather systems that are building - and predictions made by forecasters and computer 'models'. For wind direction, for instance, these 'models' work by taking a snapshot of current airflow around the world to project what could happen the next day or the day after - based on the state of the atmosphere and taking into account the amount of energy.
Another problem is that slight changes in weak winds are picked up less easily than big ones, so errors could result. And while technological advances have made predictions more reliable and accurate than they were, say, 10 years ago, being in the tropics presents challenges, said the NEA.
'The weather systems in the tropics are weaker and smaller. The winds, for example, are generally lighter, and the thunderstorms are smaller in size, and small changes could slip through the observation net. 'That is why reliability of predictions is reduced to two to three days, not a week like in colder latitudes,'' Mr Lam explained.
There is also currently a 'weak' El Nino effect, he said. El Nino is the weather phenomenon that sparked the 1997 haze, and could cause dry weather in Sumatra and Borneo. According to Associate Professor David Higgins, from the Department of Geography at the National University of Singapore, that could mean less rain than usual.
Most haze watchers are banking on the anticipated wet weather to help extinguish some of the fires. Drier conditions could also spark natural forest fires, warned Dr Rajashekar Bala, an air quality specialist from NUS' Faculty of Engineering. The widespread occurrence of peat in parts of Indonesia, including south Sumatra and Kalimantan, makes the situation worse since peat easily catches fire.
Yesterday, satellite images detected 183 hot spots in Jambi and South Sumatra, and 637 in Borneo. The 24-hour PSI stayed in the moderate range, at 71. While it cleared up slightly in the morning, it deteriorated again in the afternoon when the prevailing winds changed from south-easterly to south-westerly.
Rain over Sumatra yesterday morning also helped extinguish some of the fires, it added. Still, the NEA said the dense smoke which remains over central and south Sumatra, together with winds that are not in Singapore's favour, will mean hazy conditions are likely to continue today.