With another major international conference on climate change starting in Argentina on Monday, delegates will be pondering a sombre message from a panel of respected British climatologists.

Margaret Gilmore: Floods of past week just a a taste of things to come

Scientists from the Hadley Centre on Climate Change, part of the UK's Meteorological Office, have published a new scenario of climate change.

They say that based on the temperature peaks of every year in the last millennium, 1998 is likely to end up as the UK's hottest year since 1106.

Professor Phil Jones said: "The work we have done with various proxy climate indicators would indicat that 1998 will probably be the warmest year of the millennium."

Their predictions are based on the most recent version of the centre's climate model, and assumes that the world will go on pouring out greenhouse gases without doing anything to reduce them.

The report paints a dire picture of a world many people alive today will inhabit between 2041 and 2070.

World health will suffer

It will be, they say, a world with many more sick, hungry and thirsty people, because of climate change. And they have identified a new threat which earlier predictions failed to include.

Fifty years from now, the world's forests will not be helping to soak up the main greenhouse gas, carbon dioxide(CO2). They will themselves be a significant source of carbon emissions.

The report confirms several earlier, more tentative predictions by other scientists, notably those working for the Intergovernmental Panel on Climate Change, the most authoritative worldwide group of climatologists.

Comparisons of climate model simulations and actual observations show that human greenhouse gas emissions have contributed substantially to global warming over the past half century.

Climate models can simulate reasonably well the climate change of the past 150 years, and, the Hadley Centre group says, "this gives us confidence in predictions of the future".

More extreme weather conditions are expected

There are confirmations of what is happening now. The 1997-1998 El Niño climate disturbance in the Pacific was the most extreme on record.

Looking to the future, the research backs up the worst fears of previous forecasters.

Over the next century, greenhouse gas emissions will increase warming by about 3°C - the most extreme rise in 10,000 years.

Perhaps the most serious part of the report is that dealing with the role of the world's forests.

Rainforests will be a significant source of carbon emissions.

It predicts that tropical forests in northern Brazil will die back in the 2050s, and globally tropical grassland will be transformed into desert, or at least temperate grassland.

For the first half of the 21st century, vegetation will absorb CO2 at a rate of about 2-3 billion tonnes (1 bn tonnes = 1GtC) per year. Human emissions of CO2 are about 7GtC a year.

But from 2050 onwards, vegetation dying under the impact of climate change will itself add about 2GtC a year to greenhouse emissions, further intensifying global warming.

In a masterpiece of judicious understatement, the authors say: "This enhancement is not yet included in climate predictions."

This is a diplomatic way of recognising "positive feedback" - a way by which the global warming we have caused will itself cause further and quite unpredictable damage.

And there are some quite specific predictions about what lies in store.

About 170 million people globally are predicted to suffer from extreme water shortage.

Risk of starvation to increase

Crop yields will increase in areas like Canada and Europe, but nearer the equator they will shrink. Africa will be worst affected, with 18% more of its people at risk of hunger simply because of climate change.

Global sea levels will rise by 21cms and a further 20m people will be at risk from flooding, particularly in south and south east Asia.

If the predictions are correct, malaria infection will also increase, and in areas where it is not currently endemic.

Controversy continues to rage over the reliability of climate change forecasts, and over the very notion that climate change is being caused by human activity, and not by natural cycles.

But the overwhelming consensus of climatological opinion insists that climate change is real, and that we are playing the chief part in causing it.