Climate change seems to be back on the political agenda: a majority of the U.S. public favors action, and President Obama has promised new efforts. The hard part is deciding what to do.

Previous attempts to deal with global warming have focused on marginally reducing greenhouse-gas emissions, but that doesn’t come close to solving the problem. The carbon dioxide we put into the atmosphere stays there effectively forever, so the problem will just get bigger until we stop all emissions. Our strategy should aim to eliminate them and deal with the harm we have already caused. Here’s an outline of how we could do it.

Reaching zero emissions while meeting our energy needs will require us to reduce demand. Our first step should be to commit to never building another energy-inefficient city, building, vehicle, or industry.

Next we should focus on energy production, resolving to produce only zero-emission electricity and to electrify heat and transportation. Renewable electricity is popular, but we have to overbuild by a factor of three or four to make up for times the sun doesn’t shine and the wind doesn’t blow. Alongside that extensive new infrastructure, large-scale energy storage and some ability to control demand to fit supply will be needed, and that could take many decades. Plentiful, cheap natural gas creates half the emissions of coal, but making that fuel part of a zero-emission world will require some means of carbon capture and storage (CCS). Nuclear power works 95 percent of the time at large scale without emissions, and new passively safe designs are available. We can create a reliable emission-free electricity system for both rich and poor nations by combining CCS, nuclear power, and renewables.

Not all transport can be electrified, and it is unlikely that biofuels can be made at scale without affecting food availability or generating greenhouse gases, so we must also search for alternative scalable, non-biomass-based zero-carbon fuels.

No matter how fast we act, the effects of climate change could become severe. Could we remove carbon dioxide from the atmosphere, or shade the planet with sulfate particles? (See “A Cheap and Easy Plan to Stop Global Warming”) Most such geoengineering concepts will probably turn out to be bad ideas, and caution is appropriate. But we may need to take steps like these in the future. We should begin systematic geoengineering research now.

Nuclear power, CCS, electrification, energy storage, decarbonized fuel, efficiency standards, and geoengineering: our to-do list is long. But we have no choice.