The 2010s were almost certainly the hottest decade on record — and it showed. The world burned, melted and flooded. Heat waves smashed temperature records around the globe. Glaciers lost ice at accelerating rates. Sea levels continued to swell.

At the same time, scientists have diligently worked to untangle the chaos of a rapidly warming planet.

In the past decade, scientists substantially improved their ability to draw connections between climate change and extreme weather events. They made breakthroughs in their understanding of ice sheets. They raised critical questions about the implications of Arctic warming. They honed their predictions about future climate change.

As another decade begins, scientists say there are more questions to be answered. We asked climate researchers across a variety of disciplines about the biggest priorities and hottest topics for the 2020s. Here’s what they said.

ARCTIC MYSTERIES

The Arctic is warming faster than anywhere else on Earth, with temperatures rising at least twice as fast as the global average. Many scientists believe that understanding the consequences of Arctic warming is essential for making accurate predictions about climate change around the world.

Some of these links are straightforward. Melting Arctic ice pouring into the ocean can raise global sea levels. Thawing permafrost can release large amounts of carbon dioxide and methane into the atmosphere, potentially accelerating the rate of global warming.

Others are more contentious.

In the last decade, a growing scientific debate has arisen about the influence of Arctic warming on global climate and weather patterns, particularly in the midlatitudes.

Some observational studies have pointed to a statistical connection between Arctic warming and weather events in places like the United States, Europe and parts of Asia — for instance, a link between shrinking sea ice and cold winters in Siberia, or Arctic heat waves and extreme winter weather in the United States.

The trouble is models have a hard time capturing the causes driving these connections.

“No one argues that the Arctic meltdown will affect weather patterns, the question is exactly how,” said Arctic climate expert Jennifer Francis, a researcher at Woods Hole Research Center. “So figuring out what’s not right in the models will be a major focus. Without realistic models, it’s hard to use them to separate Arctic influences from other possible factors.”

Resolving the debate will require “a combination of data and modeling,” according to NASA climatologist Claire Parkinson. Many scientists are already hard at work on this issue.

One ongoing project known as the Polar Amplification Model Intercomparison Project is conducting a series of coordinated model experiments, all using the same standard methods, to investigate the Arctic climate and its connections to the rest of the globe. Experts say these kinds of projects may help explain why modeling studies conducted by different groups with different methods don’t always get the same results.

At the same time, improving the way that physical processes are represented in Arctic climate models is also essential, according to Xiangdong Zhang, an Arctic and atmospheric scientist at the University of Alaska, Fairbanks.

Outside that debate, there are still big questions about the Arctic climate to resolve. Scientists know the Arctic is heating up at breakneck speed — but they’re still investigating all the reasons why.

Researchers believe a combination of feedback processes are probably at play. Sea ice and snow help reflect sunlight away from the Earth. As they melt away, they allow more heat to reach the surface, warming the local climate and causing even more melting to occur.

One key question for the coming decade, Zhang said in an email, is “what relative role each of the physical processes plays and how these processes work together” to drive the accelerating warming.

Unraveling these feedbacks will help scientists better predict how fast the Arctic will warm in the future, according to Francis — and how quickly they should expect its consequences to occur. They include vanishing sea ice, thawing permafrost and melting on the Greenland ice sheet.

OCEANS AND ICE

Sea-level rise is one of the most serious consequences of climate change, with the potential to displace millions of people in coastal areas around the world.

At the moment, the world’s oceans are rising at an average rate of about 3 millimeters each year. It appears to be speeding up over time. That may not sound like much, but scientists are already documenting an increase in coastal flooding in many places around the world.

Accurately predicting the pace of future sea-level rise is one of the biggest priorities in climate science. And one of the biggest uncertainties about future sea-level rise is the behavior of the Greenland and Antarctic ice sheets, both of which are pouring billions of tons of ice into the ocean each year.

Recent satellite studies have found that ice loss in both places is speeding up. Antarctica is losing about three times as much ice as it was in the 1990s, while losses in Greenland may be as much as seven times higher than they were in previous decades.

Investigating the processes driving the accelerations — and then using that knowledge to fine-tune predictions of future sea-level rise — is a key priority for 2020 and beyond, according to Marco Tedesco, an ice sheet expert at Columbia University.

“How do we connect the physical processes that we do understand are creating this acceleration from Greenland and Antarctica, very likely over the next decade, to sea-level rise impacts?” he asked E&E News. “And how do we account for the potential shocks of the things that we cannot anticipate still?”

Some scientists worry that as ice loss continues to speed up in both Greenland and Antarctica, parts of the ice sheets could eventually destabilize and collapse entirely — leading to catastrophic sea-level rise.

In recent years, scientists have discovered that warm ocean currents are helping to melt some glaciers from the bottom up, both in Greenland and particularly in parts of West Antarctica. Better understanding the relationship between oceans and ice is a key priority for glacier experts, Tedesco said.

At the same time, monitoring the way water melts and moves along the top of the ice is also a major priority. In Greenland, climate-driven changes in the behavior of large air currents like the jet stream may be helping to drive more surface melting.

“The important thing is to understand how Greenland mass loss can be connected to the recent changes in the atmospheric circulation that we are witnessing,” Tedesco said.

EXTREME WEATHER EVENTS

The past decade saw leaps and bounds in a field of climate research known as “attribution science” — the connection between climate change and extreme weather events.

It was once thought to be impossible, but scientists are now able to estimate the influence of global warming on individual events, like heat waves or hurricanes. In the past few years alone, scientists have found that some events are now occurring that would have been impossible in a world with no human-caused climate change.

As attribution science has advanced, researchers have been able to tackle increasingly complex events, like hurricanes and wildfires, which were previously too complicated to evaluate with any confidence. They’ve gotten faster, too — researchers are now able to assess some extreme events nearly in real time.

Some organizations are working to develop sophisticated attribution services, similar to weather services, which would release analyses of extreme events as soon as they occur. The German national weather service; the United Kingdom’s Met Office; and the Copernicus program, part of the European Centre for Medium-Range Weather Forecasts, have all begun exploring these kinds of projects.

At the same time, scientists are working to improve their predictions of future extreme events in a warming world.

So far, climate models predict that many extreme weather events will happen more frequently, or will become more severe, as the climate continues to change. Heat waves will be hotter, hurricanes will intensify, heavy rainfall events may happen more frequently in some places, and droughts may be longer in others.

Continuing to improve these kinds of predictions — and then communicating them in useful ways to communities that will be affected by them — is a major priority, according to Piers Forster, director of the Priestley International Centre for Climate at the University of Leeds.

There’s often great uncertainty when it comes to predicting extreme weather events, he noted — different climate models sometimes deliver vastly different results. But it can often be both expensive and time-consuming to run the models enough times, and at high enough resolutions, to investigate and narrow these uncertainties.

Tackling this issue is one of the key challenges for climate modeling in the coming years, Forster said, noting that “we need to get clever about how we use models to make projections and how we test them.”

PROJECTING THE FUTURE

Predicting how much the Earth will warm, given a certain level of greenhouse gas emissions, may seem like the simplest goal of climate modeling. But it’s harder than it sounds.

Climate models don’t always agree on the Earth’s exact sensitivity to greenhouse gas emissions — although they do tend to fall within a certain range. If global carbon dioxide concentrations were to double, for instance, models from the past decade have tended to predict that the Earth would warm from between 1.5 and 4.5 degrees Celsius.

Scientists around the world are working on a new suite of updated climate models, which will be used to inform future reports produced by the Intergovernmental Panel on Climate Change. But there’s one issue that’s already raising eyebrows, according to Zeke Hausfather, a climate scientist at the University of California, Berkeley — so far, the newer models seem to be predicting a much higher climate sensitivity than the older models.

“The high end is much higher,” he told E&E News. “There’s a number of models above 4.8 degrees sensitivity and even up to 5.6.”

Only about 20 new models have submitted results; far more will come pouring in before the project is complete. And as Hausfather pointed out, other recent studies have suggested that the Earth’s climate sensitivity might actually be narrower than the old models suggested.

But it’s something to keep in mind at a time when accurate predictions about future warming are more pressing than ever.

“The fact that some of these models are high is interesting but doesn’t necessarily mean we should believe them over other lines of evidence,” Hausfather said. “It just reflects the fact that climate sensitivity is this huge remaining source of uncertainty in our climate projections.”

At the same time, climate modelers are also working to hone their projections in other ways. Models are able to capture increasingly complex processes the more they advance. But there are still a few key areas scientists are focused on improving.

Clouds, for instance, are believed to have a significant influence on the climate system. But they’re notoriously difficult to reproduce in climate models. Certain aspects of the carbon cycle are also underrepresented in models, Hausfather noted — for instance, the way that forests and oceans absorb or release greenhouse gases into the atmosphere.

And scientists are also working to develop more realistic climate scenarios for their modeling projects. In the past, many studies have focused on a “business as usual” climate scenario, which suggests high rates of future greenhouse gas emissions, the continued expansion of coal, and other assumptions about industry and socioeconomics that may no longer be realistic, according to Hausfather.

While global climate action is still significantly lagging when it comes to meeting the goals of the Paris climate agreement, the future may not be as dire as previous business-as-usual climate studies would suggest. Focusing new studies on more realistic scenarios may be more useful to policymakers and communities trying to plan for the future.

“In many ways the range of possible futures is narrowing,” Hausfather said. “As we get closer to 2100 and as the world takes more climate action, the worst-case 4 degrees-plus warming scenarios are a lot less likely.”