Massive, futuristic schemes to spur land and sea into sucking up greenhouse gases may help the fight against global warming but are no substitute for reducing the pollution itself, scientists said Wednesday.

Once dismissed as daft or dangerous, some of these "geo-engineering" projects can be of use in fending off the juggernaut of climate change, but only if they go hand-in-hand with cuts in carbon emissions, they warned.

"Geo-engineering" describes large-scale schemes such as erecting sunshades or mirrors in space, sowing the stratosphere with white particles or whitewashing building roofs to reflect sunlight, or scattering iron filings in the ocean to promote carbon-gobbling algae.

None of these projects has been launched on any significant scale.

Green groups are deeply suspicious of them, saying the most ambitious ventures could wreck ecosystems, carry an astronomical price and postpone tough decisions on reducing emissions of fossil-fuel gases that cause the problem.

But promoters of geo-engineering are now getting a closer hearing as political efforts to resolve climate change remain bogged down.

They argue that geo-engineering, by slightly cooling the planet, would buy time for humans to get their carbon pollution under control.

In a paper published on Wednesday in the journal Atmospheric Chemistry and Physics Discussions, researchers at Britain's University of East Anglia make the first attempt at calculating the effectiveness of these schemes.

They do not analyse environmental impact, nor do they estimate the cost.

"We found that some geo-engineering options could usefully complement mitigation [of emissions], and together they could cool the climate," said Tim Lenton, a professor of environmental sciences. "But geo-engineering alone cannot solve the problem."

The report said:

-- To gain a quick cool by 2050, building a deflector in orbit and sowing the stratosphere with fine sulphur particles are the best bet. But they also carry "a heavy burden of risk." The particles have to be replenished, and the sunshade would need maintenance. Any breakdown would cause temperatures to rise at a stroke.

-- "Fertilising" the ocean to boost plankton growth has chiefly long-term potential, as it would take centuries or millennia to really get up to speed.

-- Over the shorter term, it makes more sense to plant trees to soak up atmospheric carbon dioxide, and to use energy by burning biomass by pyrolysis so that its residues are returned to the soil as charcoal, a form of carbon that remains stable for centuries or millennia.

-- Painting roofs and roads white and other actions to help land surfaces reflect solar rays is of limited and local value. It could cool cities a little, but globally would be of little effect.

-- The benefits of some geo-engineering schemes have been in exaggerated in the past, and calculations about their effectiveness are fraught with errors.

The study comes amid an intensifying debate among climate experts about geo-engineering.

In its landmark report in 2007, the UN's Intergovernmental Panel on Climate Change (IPCC) blasted geo-engineering options as "largely speculative and unproven, and with the risk of unknown side-effects."

In the most controversial experiment to date, scientists aboard a German research vessel, the Polarstern, are in the Southern Ocean where they plan to carry out what they describe as a small-scale test in iron fertilisation.

Germany's environment ministry has spoken out against the experiment, but the research ministry has stood by it.