Scientists are revising global warming projections again, telling the public that two bookends of earlier climate scenarios—the “worst” and “best” futures—no longer look like the most plausible extremes. In the update, researchers said modest progress in reducing the highest carbon pollution pathways has reduced the likelihood of the most catastrophic heating, even as they acknowledged that the world is still not on track to meet the Paris Agreement goal of 1.5 degrees Celsius without later, engineered cooling.

The work replaces two common extremes with a set of seven plausible carbon pollution pathways for the future. Researchers said the changes reflect how countries increasingly generate electricity using low-carbon sources such as solar, wind and geothermal—technologies that do not emit carbon dioxide—while also recognizing that those gains have not come quickly enough to drive down global carbon pollution fast enough to keep warming low.

Under the Paris Agreement, adopted in 2015, the goal is to limit warming to 1.5 degrees Celsius above pre-industrial levels, a benchmark that became a central slogan for climate policy. But scientists said their updated scenario set leaves no pathway that can plausibly keep warming at or below 1.5 degrees Celsius based on emissions reductions alone. They said the “best” future still passes the 1.5-degree mark.

Climate scientist Detlef Van Vuuren, of Utrecht University and lead author of the study laying out the future scenarios, said the revised worst-case scenario reaches about 3.5 degrees Celsius of warming by the end of the century. He also said the updated best-case future is a few tenths of a degree Celsius warmer than previously theorized, in a range that still edges past the Paris target.

Johan Rockström, director of the Potsdam Institute for Climate Impact Research, described the overall shift as a narrowing of plausible futures. He said, “There is kind of a narrowing of the futures. It cannot be as bad as we thought, but it cannot be as good as we hoped,” reflecting the idea that the range of outcomes is compressing even as the upper portion remains dangerous.

Scientists also said that even the “middle” pathway would keep the planet on a trajectory similar to the one society is currently following. They said Earth is now about 1.3 degrees Celsius warmer than pre-industrial conditions and that relatively small additional warming can drive major consequences for ecosystems, including species losses, increasing water scarcity, and intensifying extreme events such as floods and heat waves.

For the long-term, researchers said carbon dioxide persists in the atmosphere for roughly a century, which limits how quickly temperature trends can reverse. Climate scientist Bill Hare, CEO of Climate Analytics, said the remaining best-case future relies on warming that rises past 1.5 degrees Celsius, peaks around 1.7 degrees Celsius and then would have to come back down below 1.5 degrees if technology could eventually remove massive amounts of carbon from the air. Hare warned that the message is not just about climate models but about political failure, telling readers, “This is just physics.” He added, “We’re losing the ability to limit warming even by two degrees without strong action and people need to be aware of that and be aware that it’s a political failure. It’s not an act of God or anything. It is just because politicians in many places are not acting fast enough.”

Natalie Mahowald, a climate scientist at Cornell University and co-author of a U.N. science report on harms above 1.5 degrees Celsius, said the 1.5 target is not simply a policy number. She said the implications of not meeting it include heightened harms, and she warned that impacts would be especially severe for small island developing states, saying: “Some of them will go underwater.”

The revisions also sparked debate about how earlier high-end projections were framed. Roger Pielke Jr., of the American Enterprise Institute, said changes to the highest end scenario matter because it was presented as a likely future that could come true without changes to policies, even though he said research had already suggested it was improbable. In an email, Pielke said the scenario “It was always presented as where we were headed absent explicit climate policy,” and he said it was based on out-of-date and incorrect coal-heavy energy assumptions.

Keywan Riahi, who led the 2011 study that introduced that higher-end pathway and is director of the Energy, Climate and Environment Program at the International Institute for Applied Systems Analysis, argued that the high-end case was never intended to represent what scientists believed the world was heading toward. In remarks included in the reporting, he said, “It was never a likely case. It was basically, given the underlying studies in the literature at that time, a plausible higher bound of what possible emissions could look like. This is very different than if you would ask the question, what is now the most likely scenario.” He added that “It’s a success story,” describing declines in the cost of renewables, including solar and wind, as evidence that the high-end path did not become reality.

President Donald Trump also seized on the debate in a social media post that characterized earlier projections as wrong, writing, “GOOD RIDDANCE! After 15 years of Dumocrats promising that ‘Climate Change’ is going to destroy the Planet, the United Nations TOP Climate Committee just admitted that its own projections (RCP8.5) were WRONG! WRONG! WRONG!” Van Vuuren responded to that framing by emphasizing that risks remain even if the most dramatic pathway did not occur, telling readers, “The risks of climate change have not disappeared.” He said, “The good news is that we did not follow the most dramatic emission pathway. However, we are still heading towards a future with significant climate impacts; a future we should avoid.”

Even with emissions pathways narrowed, scientists said an important uncertainty remains. Mahowald, Rockström and Hare said the latest scenarios focus on carbon from fossil fuel burning—the primary emissions lever humans can control—while acknowledging that other “feedback” processes in the climate system can add additional warming on top of emissions-driven heat. Rockström said those feedbacks include carbon stored in oceans and forests, changes to ocean currents, and cloud reflectivity, and he said these processes can be difficult to project, with some scenarios allowing an extra half degree Celsius or nearly a degree Fahrenheit beyond what emissions alone would produce.