THE MERCURY IS bubbling at close to 31 degrees Celsius on the thermometer under a thatched roof on the lawn outside Hong Kong Observatory in Tsim Sha Tsui. The temperature gauge is recording what is destined to be one of the hottest and sunniest Julys since the colonnaded Observatory building was constructed in the 1880s. Ten minutes later the blue sky gives way to a dark cloud that unleashes a torrential downpour. The weather is a fickle creature and, the advances of meteorological science notwithstanding, remains frustratingly hard to predict. It is a fact that the Observatory's assistant director Yeung Kai-hing acknowledges with a smile although, he says, there are some predictions that can be made with certainty. One of them is that Hong Kong is getting hotter. The 1990s was the hottest decade since records began in 1884, and this decade may see the record books scorched again. Seven of Hong Kong's eight hottest years have occurred since 1990. The hottest year so far was 1998 with a year-round average temperature of 24 degrees Celsius. Last year ran a close second, with a mean temperature of 23.9 degrees. The first six months of this year rank No 8 on the all-time scale, but with a steaming summer could rise up the table by year's end. 'Hong Kong is getting warmer and will continue to do so,' says Yeung. Questions about how much warmer and what is causing it are more difficult to answer. What is known is that during the past 50 years, Hong Kong's temperature has risen by about one degree Celsius and the big upswing since 1990 suggests the average could rise further in future. It is not a local phenomenon. On July 2, the normally staid statistical body the World Meteorological Organisation (WMO) issued an unprecedented alert about climate change. 'Analyses of proxy data for the northern hemisphere indicate that the increase in temperature in the 20th century is likely to have been the largest in any century during the past 1,000 years,' it said. It added that in the northern hemisphere the 1990s was the warmest decade and 1998 the warmest year, mirroring Hong Kong's experience. Globally, this year could be the hottest ever. The WMO also pointed to a huge upswing in the past 30 years. The trend for warmer temperatures since 1976 was roughly three times that for the past 100 years as a whole. Global average land and sea surface temperatures in May 2003 were the second highest since records began in 1880. The WMO warned that the climate change could lead to an increase in the number and intensity of extreme weather events such as droughts, floods and record high - as well as record low - temperatures. It cited evidence that the world's weather was already going haywire. In southern France, record high temperatures were recorded last month - soaring above 40 degrees Celsius, between five and seven degrees above the long-term average. Switzerland sweltered under its hottest June for at least 250 years. In the United States, there were 562 tornados during May, a record for any month and almost 50 per cent more than the previous high of 399 tornados in June 1992. In India, this year's pre-monsoon heatwave brought peak temperatures of between 45 and 49 degrees, up two to five degrees on the long-term average and claiming 1,400 lives. In Sri Lanka, tropical cyclone 01B caused flooding and landslides that killed 300 people and decimated the tea harvest. 'New record extreme events occur every year somewhere in the globe, but in recent years the number and intensity of such extremes have been increasing,' the WMO warned. Yeung says the WMO's pronouncement is not based on 'ad-hoc research', but reflects the findings of systematic studies carried out by the Intergovernmental Panel On Climate Change, an international weather watchdog set up jointly with the United Nations Environmental Programme. 'These changes in climate do not occur uniformly across the globe,' says Yeung, noting that countries lying in the Earth's mid- to-high latitudes are most affected. 'In this part of the world the impact is less.' Nevertheless, the Observatory's figures reveal Hong Kong's experience is following the global trend. Yeung says the rise is down to two factors: urbanisation and global warming. Which factor is dominant is hard to say because of a lack of comparable data. 'We may not be able to say with certainty that this much is due to urbanisation and this much due to global warming, but it's possible to tell we are affected by both,' says Yeung. Global warming affects the minimum temperature more than the maximum because the so-called greenhouse gases released into the atmosphere act as a blanket trapping the Earth's heat. 'It traps the infra-red radiation thus slowing temperature falls at night,' says Yeung. This means the coldest night-time temperatures will be warmer than in the past. Figures bear this out; the 10 highest minimum daily temperatures have all occurred since 1990, with the mercury never dropping below 29.3 degrees on those days. The 10 lowest minimum temperatures all occurred before 1956. While temperatures in rural areas can be up to 10 degrees cooler than in urban areas, the average difference is not so stark. Weather stations at remote outposts such as Waglan Island, to the south east of Hong Kong, have shown less marked increases temperatures since it was installed 40 years ago, which suggests urbanisation is the most important influence. Yeung says this is not necessarily the case because recent research has found the cooling influence of the ocean on coastal areas is significant. 'Greenhouse gases tend to affect land more than the ocean; land heats up faster,' says Yeung. But he warns the sea temperature will inevitably follow suit. Not everyone agrees that the world's temperature will go on rising, or that the cause is man-made. 'There is no question the world is warming up, everyone agrees on that,' says weather watcher Johnny Chan Chung-leung, chair professor of applied physics at City University. 'The question is whether global warming is due to natural variability or human-induced? I am not sure what the mechanism is.' The world's rainfall is often attributed to the extent of the El Nino phenomenon, a disruption of the ocean-atmosphere system in the tropical Pacific. Yeung pins the blame on El Nino for Hong Kong's wettest year in 1997 and its fourth-wettest in 2001. Chan argues that temperature chan- ges may occur in natural cycles, with heating or cooling patterns renewing perhaps every 20 or 50 years. 'We need to go back to the history books and see whether the natural variability has the same period of oscillation,' he says. 'We haven't done that. If this warming were totally contributed by anthropogenic [man-made] forces then we would expect it to increase, but if natural variability is large and global cooling occurs, they could cancel each other out,' he says. Chan says a battle between natural and man-made causes of climate change will determine future temperatures. Yeung says the Observatory's computers are not big enough to run accurate simulation programmes for the next 50 or 100 years. But megacomputers in Europe, Japan and the United States all predict temperatures will continue to rise 1.4 degrees Celsius to 5.8 degrees Celsius by the end of this century, depending on green-house gas emissions. As the rain stops and the sun again beats down from a clear blue sky, Yeung takes his time when pushed to estimate Hong Kong's temperature rise over the next 50 years based as much on his intuition as scientific data. 'Perhaps another one or two degrees,' he says. 'But we know it will be warmer.'