This year, Las Vegas, Nevada broke its all-time heat record, reaching 120° F.

The temperature was recorded at Harry Reid International Airport on July 7, 2024. That week, between July 6 and July 12, was the new hottest 7-day period on record, with an average high temperature of 117.5° F.

This is the daily reality for Vegas residents in the summer. Record-breaking temperatures are hard to bear, but so were all the hot days and nights that came before. Commuters frequently see temperatures above 120 flash on their vehicle dashboards, and outdoor workers struggle to do their daily tasks under the hot sun.

“There’s a disconnect between climate science and the people who live here,” says Woodwell Climate Research Associate, Monica Caparas. “Vegas residents know our summers are hot and unbearable. Understanding climate change is driving the extreme weather we’re experiencing is where the disconnect lies. ”

Caparas moved to Las Vegas as a child. She grew up there, left for college, and returned to settle into her adult life. Today, she works for Woodwell Climate’s Risk team remotely from her home in the city. Caparas knows the ins and outs of local life. These include Vegas’s rapid population expansion, the groups of people experiencing homelessness sheltering in underground stormwater infrastructure, and the heat that was unbearable before it started making headlines.

Experiencing climate change without shelter

Caparas’s work with the Risk team aims to provide communities like Las Vegas with an accurate picture of the climate-driven changes in their future. These “risk assessments” are provided through Woodwell Climate’s Just Access program, which uses the most accurate climate models, in collaboration with local knowledge, to anticipate future community safety threats. The analyses have brought to light growing threats from flooding, heat, storms, and more. The team provides assessments, free of charge, to states, cities, and countries across the world.

Just Access serves what Risk Program Director Christopher Schwalm calls “frontline communities.” The term describes groups of people who are over-exposed, under-resourced, underserved, historically marginalized, and therefore the most at-risk to the repercussions of climate change. In the risk assessment for Las Vegas, people experiencing homelessness are front and center.

“Between May 20th and the first week in July, about 20 people who were experiencing homelessness died of heat,” says Dr. Catrina Grigsby-Thedford, Executive Director of the Nevada Homeless Alliance (NHA) and community partner in Las Vegas.

The NHA estimates that almost 8,000 people are experiencing homelessness on any given night in southern Nevada. The number is only growing. Grigsby-Thedford says that this year’s unhoused population is up 1,300 people compared to 2023.

“Often our shelters are full,” Grigsby-Thedford says, “We’re limited by shelter beds and space.”

The NHA’s shelters do open all day in extreme heat, but so many people packed tightly together is still unsafe.

With nowhere to go, some seek shelter underground in Las Vegas’s stormwater infrastructure. While the tunnels are cooler out of the sun’s reach, they are at risk from flooding. Across the region, extreme precipitation is expected to increase by 12-14% by 2050, raising flood risk in the city and especially within the tunnels.

To combat lack of space and shelter, the NHA hosts 4-8 one-stop resource fairs per month. The events, called Project Homeless Connect, serve both people experiencing homelessness and low-income residents in Las Vegas. Grigsby-Thedford says these events “fill in gaps”—offering housing assistance, medical care, hygiene care, and other resources.

Despite all of this work, many unhoused people are hesitant to engage with organizations like the NHA. Grigsby-Thedford says “choice is often a challenge,” and that when people grow accustomed to the way things are, they often accept it and choose to stay.

Picturing risk

Building trust with communities, especially those predisposed to mistrust outside actors, is essential in this work. Which is why, Schwalm says, Woodwell Climate approaches risk work with the goal of “meet[ing] people where they are.”

That means “scoping,” the team’s word for listening to what community and government leaders want out of the risk analysis—what concerns they have, weak points they’ve identified, and what help might be needed post-analysis.

“Two-thirds of the time we spend from start to finish falls into this scoping idea, rather than doing analysis itself,” Schwalm says.

Scoping frames the data the risk team collects, as well as who their partners will be during the risk analysis process.

“We find people who are practical and recognize that there’s a problem,” Schwalm says, “We only work with communities who want to work with us.”

Following the scoping process, the Risk team compiles an analysis of extreme weather events and subsequent risks each community will face as climate change progresses.

“We perform a stress test of that particular geography to identify weak points,” Schwalm explains.

Then, the Risk team uses the most up-to-date climate models possible to predict changes in extreme weather and regional climate. By using predictive models, the team focuses efforts on what the future will hold, as opposed to using past strategies.

“We need to use the future to predict the future,” Schwalm says simply.

Making climate risk data accessible to all

Over the past three years, Just Access has provided 50 communities—that’s about a quarter billion people—with risk analyses. These communities span the U.S., Central and South America, Africa, Asia, and Oceania. They’ve worked with countries, like the Democratic Republic of Congo, where they helped update the country’s National Adaptive Plan, states like Chiapas in Mexico, groups like Cree Nation in Canada, and other communities, now including Las Vegas.

Despite all of this work, though, Schwalm says there is still room to grow.

“Fifty communities is kind of only a drop in the bucket,” he says, “We’re not going to make a huge dent in this unless we move beyond working community-by-community.”

Two major roadblocks for Just Access are finite resources: time and money. Individual risk analyses require a lot of time and communication to address risks in relatively small areas.

The other obstacle, money, is something climate research could always use more of. Grants and donations are crucial in order for analyses to remain free, and those sometimes come with limitations.

“There’s a tension from the funder to work in a specific geography sometimes,” Schwalm says, “It’s a juggling act.”

Climate change can also be a politicized topic. In order to meet people where they are, sometimes the Risk team implements changes in language used to communicate with community leaders. This can be a change as simple as using “extreme weather” instead of “climate change.” As long as everyone in the room is ready to confront what the future holds, they’re all working on the same page towards the same goal.

“We’ve done red states, blue states, rural, urban,” Schwalm continues. “We’ve learned how to read the room.”

Creating the foundations for change

Woodwell Climate’s involvement in Las Vegas brings to light the way justice issues, like homelessness, interact with growing threats from climate change.

“In the Las Vegas risk assessment, we are focusing on the disproportionate impacts of the climate crisis on communities already facing systemic socio-economic inequity,” says Caparas. “We must think about intersectionality in order to address climate justice.”

Not only does climate change represent a current crisis for those experiencing homelessness, communities with fewer resources are now at greater risk of being made homeless by future climate-related disasters. Accurate climate risk information can support organizations like NHA as they develop strategies to serve people experiencing homelessness in a more extreme future.

Grigsby-Thedford says that NHA members, especially those with lived experience of homelessness who work as Lived X Consultants, are always looking to be involved in projects like the one Caparas leads.

“We always talk about weather in our meetings,” she says, “So this is perfect, someone’s actually doing research about this. Anything that impacts [Las Vegas’s homeless population], we want to make sure we’re involved in that.”

For the Las Vegas risk assessment, Caparas is working with the NHA and Southern Nevada Lived X Consultants to understand climate risks around cooling stations in public buildings, which are a vital, air-conditioned shelter when the heat index is too high. Grigsby-Thedford says there were many more cooling stations in 2023 and 2024 compared to previous years.

Caparas also forged a connection with Miguel Dávila Uzcátegui, Southern Nevada’s Regional Transportation Commission (RTC) Senior Planner and board member of Help Hope Home. Together, they are developing a database of flooding infrastructure and updating the city’s flooding model with future climate projections. The RTC will integrate the Risk team’s model into regional planning work, updating Las Vegas’s flooding and transportation infrastructure for community safety.

None of this work would have been possible without Caparas’s diligent bridge building between the scientific resources of Woodwell Climate and the needs of people in her own community. Those connections allow science to be informed first and foremost by those most affected by climate change.

“The people closest to the problem are the people closest to the solution,” says Grigsby-Thedford.

A new study, published in Nature Communications Earth and Environment and co-authored by researchers at Atmospheric and Environmental Research, Inc. (AER) and Woodwell Climate Research Center, finds that abnormally warm temperatures in the Arctic are associated with a higher likelihood of severe winter weather including cold-air outbreaks and heavy snowfall in Northern Hemisphere continents.

“When the Arctic atmosphere is warmer than normal, we see a much higher likelihood of extreme winter weather across much of Canada, the northern U.S. and northern Eurasia,” remarked lead author, Dr. Judah Cohen at AER. “The relationship is especially strong in the northeastern sections of the continents.”

“Even though we’re seeing cold records being broken less often as the globe warms, we’ll still see debilitating spells of severe winter weather,” added co-author Dr. Jennifer Francis at Woodwell Climate. “There will be plenty of ice, snow, and frigid air in the Arctic winter for decades to come, and that cold can be displaced southward into heavily populated regions during Arctic heat waves.”

Recent disruptive extreme winter weather events—such as the deadly Texas cold spell of February 2021—have occurred and will continue to occur in the future, wreaking havoc on infrastructure, human wellbeing, and ecosystems, especially in areas unaccustomed to and ill-equipped for dealing with winter extremes.

“The Arctic may seem irrelevant and far away to most folks, but our findings say the profound changes there are affecting billions of people around the Northern Hemisphere,” added Dr. Francis. To reverse these trends, “it will take bold and rapid actions to reduce our burning of fossil fuels and the build-up of heat-trapping gasses in the atmosphere, but the tools exist if we can muster the will.”

According to Francis, recent studies have theorized that rapid Arctic warming, a pace three-to-four times faster than the globe as a whole, may increase the likelihood of extreme weather events owing to a reduced north/south temperature difference. In addition, slower westerly winds of the jet stream lead to more frequent convoluted jet-stream configurations, which lead to unusual weather.

“Disruptions in the typically stable stratospheric polar vortex may also occur more often in a warming climate,” noted Cohen, “and we know hazardous winter weather is more likely during these disruptions.”

Hurricane season in North America is underway. Already, the second storm of the year to earn a name, Beryl, has cut a destructive swath across the Caribbean and the United States. This year, the National Oceanic and Atmospheric Administration (NOAA) forecasted an extremely active hurricane season, anticipating between 17-25 named storms (the average is 14) and 4-7 major storms (average is 3). Major storms are category 3 and above with wind speeds exceeding 111 mph.

Intense seasons like this are likely to be a more common occurrence in a warmer world, as higher temperatures, rising seas, and changing weather patterns create the conditions for bigger, more destructive, longer lived, and more rapidly strengthening storms. Here’s how climate change is affecting the Atlantic hurricane season:

1. Higher temperatures mean more energy to form hurricanes

To understand how hurricanes are being affected by climate change, it’s important to understand how hurricanes are formed. They are essentially clusters of thunderstorms, building strength as they sweep westward using the energy from warm tropical waters. Under the right conditions, the Earth’s rotation will cause the cluster to spin into a cyclone shape. Because heat is energy, increases in sea surface temperatures play a critical role in strengthening these storms.

The ocean is a major heat sink for the planet, absorbing over 90% of the excess heat trapped by greenhouse gasses in the Earth’s atmosphere over the past few decades. Global sea surface temperatures have increased approximately 2.8F since the beginning of the 20th century, and ocean heatwaves — large areas of above-normal temperatures that can last for months-– are much more common and widespread. A hotter ocean means there is more energy available to fuel tropical storms, ultimately making it a more destructive event when it hits land.

2. The hotter the air, the more water it can hold

The second thing a hurricane needs to form is moisture. Water is evaporated and pulled up into the developing storm as it spins across warm waters of the tropical Atlantic. Hotter air temperatures mean more moisture can be held as vapor in the atmosphere, which allows storms to ingest greater amounts of water that will eventually condense into clouds and be released as rainfall. Condensation also releases heat into the storm, fueling its intensification. Models estimate that human-caused global warming has increased hurricane extreme hourly rainfall rates by 11%.

3. ENSO fluctuations are becoming more extreme

Climate change is also contributing to larger swings between the two phases of the El Niño Southern Oscillation (ENSO)—meaning stronger versions of both El Niño or La Niña patterns. Currently, the Atlantic is headed towards a La Niña, which favors hurricane formation because it lessens vertical wind shear. Differences in wind speeds at different heights in the atmosphere can tear a storm apart, while less shear (more consistency in wind speeds between altitudes) allows storms to stay together and build strength.

4. Tropical storms are undergoing rapid intensification more frequently

All these factors add up to more intense tropical storms in a world altered by climate change— meaning more category 3-5 storms and more big storms back-to-back. Since 1975 the number of category 4-5 cyclones has roughly doubled.

This doesn’t necessarily mean that there will be more hurricanes; however, the ones that do form can be bigger and cause more damage (on top of the already estimated $2.6 trillion in damages since 1980.) If anything, data shows a slight decrease in the number of storms, moving more slowly along their path, releasing their extreme wind and rain over a single location for longer periods.

5. Rising sea levels are making hurricanes more deadly

Sea level rise due to climate change has also made hurricanes a more dangerous threat for more people. As sea levels rise, coastlines are put at increased risk of flooding.

Sea levels have risen roughly 8 in since the late 19th century, and the rate of rise is accelerating as climate change worsens. When a hurricane makes landfall, water is pushed inland by high-speed winds in an event known as storm surge. Every additional inch of sea level rise allows the surge to travel farther inland, threatening a wider area and causing more damage, death, and injury— especially in areas where human development along the coast has exposed people and homes to greater risk.

As temperatures continue to rise, communities along the East and Gulf coasts can expect to be hit harder by destructive storms. Despite this, more and more people are choosing to live and build along the coasts, increasing the cost of damages when hurricanes do strike. Slowing warming temperatures and building adaptation measures to protect coastal communities will become more urgent as Atlantic hurricanes intensify.

We can all agree 2023 was a weird year for weather, right? The United States set a record for the number of billion dollar weather disasters. A major Amazon River tributary reached its lowest water levels in a century during extreme drought. Extreme rain in Libya caused two dams to break, destroying homes and killing over 4,000 people.

And then, of course, there was the heat. 2023 was the hottest year on record. Countries around the world saw heat records fall month after month. The Arctic was hot. The ocean was hot. And debates swirl on about whether we’ve already passed critical warming thresholds.

So how do we put 2023 in context of the greater trend of warming? Here’s what some of Woodwell Climate’s scientists have to say about last year’s record-breaking events.

Did the models predict this?

The dramatic scenes of heat and extreme weather last year prompted many to ask why temperatures had seemingly spiked way above the trend line. Was this unexpected? Was it out of the range of what scientists had modeled? Woodwell Senior Scientist, Dr. Jennifer Francis says not entirely.

“Almost exactly a year ago,” says Francis, “we had just come out of three years of La Niñas and we came close to breaking global temperature records then, even though La Niñas tend to be cooler than neutral or El Niño years. And then along came the strong El Niño of 2023.”

El Niño and La Niña are two extremes of a natural phenomenon that impacts weather patterns across the Pacific, and around the world. In an El Niño year, the prevailing trade winds that normally push warmer waters into the western tropical Pacific—allowing cooler water to well up along the western coast of the Americas—are reversed, resulting in hotter ocean surface temperatures in the eastern equatorial Pacific. When the ocean is hotter than the air above it, that heat is released into the atmosphere, often making El Niño years record breaking ones for global temperatures.

“Last year’s spike looks a lot like the last big El Niño event in 2015-2016. It’s just that now the whole system is warmer. So to me, it wasn’t at all a surprise that we smashed the global temperature record in 2023,” says Francis.

The spike put global temperatures far above the average of climate model simulations, but that doesn’t mean the models didn’t account for it. Risk Program Associate Director, Dr. Zach Zobel, says that averages tend to smooth out natural year-to-year fluctuations, when in fact the upper and lower ranges of model predictions do encompass temperatures like the ones seen in 2023.

“It was well within the margin of error that you would expect for natural variations,” says Zobel.

How does ocean heat impact the climate?

One element of last year’s heat, one that wasn’t necessarily forecasted, was the simultaneous appearance of several ocean heat waves around the globe. The ocean absorbs the vast majority of heat trapped by greenhouse gasses, and that heat can be released under the right conditions. El Niño is one example, but in 2023 it coincided with other not-so-natural marine heat waves across the world.

“In pretty much every single ocean right now there are heat waves happening, which is something quite new,” says Francis.

A couple of dynamics could be driving this. One possibility is that, after three years of La Niñas, in which equatorial Pacific ocean temperatures were generally cooler than the air, the ocean simply absorbed a lot of heat, which was then primed to be released in an El Niño year. Another, Zobel suggests, could be recent shipping laws that required shipping vessels to eliminate sulfate emissions by 2023. Sulfates are a pollutant that may have been helping bounce back solar radiation, hiding the true extent of warming.

“Usually when there’s an El Niño, the eastern tropical Pacific is very warm, but it doesn’t actually drive up ocean temperatures everywhere,” says Zobel. “That was the biggest surprise to me: how warm the northern hemisphere of the Atlantic and Pacific were for most of last year and into 2024.”

Ocean heat waves are typically long-lived phenomena, lasting many months, and so can be a useful tool for meteorologists looking to predict 2024’s extreme weather events.

“The good news is that it provides some kind of long-term predictability about weather patterns in the upcoming year,” says Francis. “The bad news is that they tend to be unusual weather patterns, because those ocean heat waves aren’t usually there.”

Will next year be hotter?

So are we in for another, hotter year after this one? Risk Program Director Dr. Christopher Schwalm says it’s likely.

“Warming predictions for 2024 from leading scientists all forecast a higher level of warming this year than last year,” says Schwalm.

Already, March 2024, was the 10th month in a row to break temperature records. Zobel says it’s typical for the year following an El Niño peak to maintain high temperatures.

“Because the ocean spent a good amount of the year last year warmer than average, that energy is typically dispersed throughout the globe in the following year,” says Zobel. “So even though the tropical Pacific might return to normal, that energy is still in the system.”

However, atmospheric scientists are already seeing signs that El Niño is slowing down and flipping to its counterpart, La Niña, adding another layer of complexity to predictions for 2024.

“The 2024 hurricane season is a large concern,” says Zobel. “La Niña is a lot more conducive to tropical cyclone development. If we combine above average numbers with the amount of energy that storms have to feed on, it’ll be a shock to the system.”

What does this mean for 1.5?

In the discussions around 2023’s temperatures, one number dominates the conversation: 1.5 degrees C. This is the amount of warming countries around the world agreed to try to avoid surpassing, in accordance with the United Nations’ 2015 Paris Climate Agreement. Estimates from Berkeley Earth say that 2023 may have been the first year spent above that threshold.

This assertion may take several years to verify— one year spent physically above 1.5 degrees of warming does not indicate the UN threshold has been permanently passed. What scientists are looking for is a clear average trend line rising above 1.5 degrees C without coming back down, and for that you need several years of data. That, regrettably, creates a lag time between climate impacts and updating climate policy. But, for many, the debate around the arbitrary 1.5 degree goal has become a distraction. Schwalm says scientists and policy-makers should be focusing on urgently combating climate change whatever the numbers say.

“We are already living in a post-Paris Agreement reality,” says Schwalm. “The sooner we admit that and reimagine climate policy, the better.”

“Actual real world impacts are going to be there, whether we’re at 1.48 or 1.52,” says Zobel.

And Francis agrees. “There are so many indicators telling us that big changes are underfoot, that we are experiencing major climate change, but reaching 1.5 isn’t going to all of a sudden make those things worse. It’s just one more reminder we’re still on the wrong track and we’d better hurry up and do something.”

Last month was hottest February ever recorded. It’s the ninth-straight broken record

snow melts next to a road, surrounded by evergreen trees

For the ninth straight month, Earth has obliterated global heat records — with February, the winter as a whole and the world’s oceans setting new high-temperature marks, according to the European Union climate agency Copernicus.

The latest record-breaking in this climate change-fueled global hot streak includes sea surface temperatures that weren’t just the hottest for February, but eclipsed any month on record, soaring past August 2023’s mark and still rising at the end of the month. And February, as well the previous two winter months, soared well past the internationally set threshold for long-term warming, Copernicus reported Wednesday.

Continue reading on Associated Press News.

Study pinpoints links between melting Arctic ice and summertime extreme weather in Europe

New research shows how last year’s warming melted ice in Greenland that increased flows of fresh, cold water into the North Atlantic, upsetting ocean currents in ways that lead to atmospheric changes.

Arctic ice floes

The Arctic Ocean is mostly enclosed by the coldest parts of the Northern Hemisphere’s continents, ringed in by Siberia, Alaska and the Canadian Arctic, with only a small opening to the Pacific through the Bering Strait, and some narrow channels through the labyrinth of Canada’s Arctic archipelago.

But east of Greenland, there’s a stretch of open water about 1,300 miles across where the Arctic can pour its icy heart out to the North Atlantic. Those flows include increasing surges of cold and fresh water from melted ice, and a new study in the journal Weather and Climate Dynamics shows how those pulses can set off a chain reaction from the ocean to the atmosphere that ends up causing summer heatwaves and droughts in Europe.

Record-breaking heat in 2023

In North America alone, 78 all time records for hottest temperature were broken over the course of June, July and August. In New Iberia, Louisiana, the temperature record was broken four times, peaking at 109 degrees F. Places as far north as Wainwright Airport in Alaska saw temperatures as high as 84 degrees.

Humidity makes the heat deadly

Extreme heat events like these present a serious danger to human health. That threat is multiplied when instances of high temperature coincide with high humidity— interrupting the ability of the human body to cool off through evaporating sweat. A recent paper, co-authored by Woodwell Climate Risk Program director, Dr. Christopher Schwalm, defines “lethal heat” as a wet bulb temperature (a measure combining heat and humidity) of 35 degrees C (95 degrees F). Prolonged exposure— over 6 hours— to temperatures exceeding this can result in death even for a healthy person keeping hydrated in the shade

According to the paper, instances of deadly heat waves are increasing with climate change. Already, with over a degree of warming, parts of Northern India are seeing annual heat events. By just two degrees of warming— a milestone we are currently on track to hit by mid-century— a quarter of the world is expected to experience a lethal heat event at least once in a decade. A significant subset of the world, particularly regions of India, Africa, South America, and the Southeastern US, can expect deadly heat conditions at least once a year at that point, and the area will expand wider with each half degree of warming.

It’s a forecast that highlights the urgency of acting to mitigate warming and developing local and regional strategies to prepare communities to handle high heat and humidity events when they do come.

“It puts this past year’s heat waves into somber perspective,” says Dr. Schwalm. “Without action, we put a lot more, potentially billions, of people at risk of heat stress or death on an annual basis. It’s a significant public health concern.”

“It’s been around a long time, actually,” muses Senior Scientist, Dr. Jennifer Francis. “It’s gotten more sophisticated, sure, and a lot of the applications are new. But the concept of artificial intelligence is not.”

Dr. Francis has been working with it for almost two decades, in fact. Although, back when she started working with a research tool called “neural networks,” they were less widely known in climate science and weren’t generally referred to as artificial intelligence.

But recently, AI seems to have come suddenly out of the woodwork, infusing nearly every field of research, analysis, and communication. Climate science is no exception. From mapping thawing Arctic tundra, to tracking atmospheric variation, and even transcribing audio interviews into text for use in this story, AI in varying forms is woven into the framework of how Woodwell Climate creates new knowledge.

AI helps climate scientists track trends and patterns

The umbrella term of artificial intelligence encompasses a diverse set of tools that can be trained to do tasks as diverse as imitating human language (à la ChatGPT), playing chess, categorizing images, solving puzzles, and even restoring damaged ancient texts.

Dr. Francis uses AI to study variations in atmospheric conditions, most recently weather whiplash events— when one stable weather pattern suddenly snaps to a very different one (think months-long drought in the west disrupted by torrential rain). Her particular method is called self-organizing maps which, as the name suggests, automatically generates a matrix of maps showing atmospheric data organized so Dr. Francis can detect these sudden snapping patterns.

“This method is perfect for what we’re looking for because it removes the human biases. We can feed it daily maps of, say, what the jetstream looks like, and then the neural network finds characteristic patterns and tells us exactly which days the atmosphere is similar to each pattern. There are no assumptions,” says Dr. Francis.

This aptitude for pattern recognition is a core function of many types of neural networks. In the Arctic program, AI is used to churn through thousands of satellite images to detect patterns that indicate specific features in the landscape using a technique originally honed for use in the medical industry to read CT scan images.

Data science specialist, Dr. Yili Yang, uses AI models trained to identify features called retrogressive thaw slumps (RTS) in permafrost-rich regions of the Arctic. Thaw slumps form in response to subsiding permafrost and can be indicators of greater thawing on the landscape, but they are hard to identify in images.

“Finding one RTS is like finding a single building in a city,” Dr. Yang says. It’s time consuming, and it really helps if you already know what you’re looking for. Their trained neural network can pick the features out of high-resolution satellite imagery with fairly high accuracy.

Research Assistant Andrew Mullen uses a similar tool to find and map millions of small water bodies across the Arctic. A neural network generated a dataset of these lakes and ponds so that Mullen and other researchers could track seasonal changes in their area.

And there are opportunities to use AI not just for the data creation side of research, but trend analysis as well. Associate Scientist Dr. Anna Liljedahl leads the Permafrost Discovery Gateway project which used neural networks to create a pan-Arctic map of ice wedge polygons—another feature that indicates ice-rich permafrost in the ground below and, if altered over time, could suggest permafrost thaw.

“Our future goals for the Gateway would utilize new AI models to identify trends or patterns or relationships between ice wedge polygons and elevation, soil or climate data,” says Dr. Liljedahl.

How do neural networks work?

The projects above are examples of neural-network-based AI. But how do they actually work?

The comparison to human brains is apt. The networks are composed of interconnected, mathematical components called “neurons.” Also like a brain, the system is a web of billions upon billions of these neurons. Each neuron carries a fragment of information into the next, and the way those neurons are organized determines the kind of tasks the model can be trained to do.

“How AI models are built is based on a really simple structure—but a ton of these really simple structures stacked on top of each other. This makes them complex and highly capable of accomplishing different tasks,” says Mullen.

In order to accomplish these highly specific tasks, the model has to be trained. Training involves feeding the AI input data, and then telling it what the correct output should look like. The process is called supervised learning, and it’s functionally similar to teaching a student by showing it the correct answers to the quiz ahead of time, then testing them, and repeating this cycle over and over until they can reliably ace each test.

In the case of Dr. Yang’s work, the model was trained using input satellite images of the Arctic tundra with known retrogressive thaw slump features. The model outputs possible thaw slumps which are then compared to the RTS labels hand-drawn by Research Assistant Tiffany Windholz. It then assesses the similarity between the prediction and the true slump, and automatically adjusts its billions of neurons to improve the similarity. Do this a thousand times and the internal structure of the AI starts to learn what to look for in an image. Sharp change in elevation? Destroyed vegetation and no pond? Right geometry? That’s a potential thaw slump.

Just as it would be impossible to pull out any single neuron from a human brain and determine its function, the complexity of a neural network makes the internal workings of AI difficult to detail—Mullen calls it a “black box”—but with a large enough training set you can refine the output without ever having to worry about the internal workings of the machine.

Speeding up and scaling up

Despite its reputation in pop culture, and the uncannily human way these algorithms can learn, AI models are not replacing human researchers. In their present form, neural networks aren’t capable of constructing novel ideas from the information they receive—a defining characteristic of human intelligence. The information that comes out of them is limited by the information they were trained on, in both scope and accuracy.

But once a model is trained with enough accurate data, it can perform in seconds a task that might take a human half an hour. Multiply that across a dataset of 10,000 individual images and it can condense months of image processing into a few hours. And that’s where neural networks become crucial for climate research.

“They’re able to do that tedious, somewhat simple work really fast,” Mullen says. “Which allows us to do more science and focus on the bigger picture.”

Dr. Francis adds, “they can also elucidate patterns and connections that humans can’t see by gazing at thousands of maps or images.”

Another superpower of these AI models is their capability for generalization. Train a model to recognize ponds or ice wedges or thaw slumps with enough representative images and you can use it to identify the water bodies across the Arctic—even in places that would be hard to reach for field data collection.

All these qualities dramatically speed up the pace of research, which is critical as the pace of climate change itself accelerates. The faster scientists can analyze and understand changes in our environment, the better we’ll be able to predict, adapt to, and maybe lessen the impacts to come.