“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.

Canada’s fire season has barely started and it’s already on track to break records. So far, NOAA has documented more than 2,000 wildfires that have resulted in the forced evacuation of over 100,000 people across Canada. The most recent bout of fires burning in Ontario and Quebec has sent smoke southward into the Eastern U.S., causing record levels of air pollution in New York and warnings against outside activity as far south as Virginia.

Only a little over a month into the wildfire season, fires have already burned 13 times more land area than the 110-year average for this time of year, and they show no sign of stopping, according to Canadian publication The Star. Indigenous communities, some of whom live year-round in remote bush cabins, have been particularly harmed by the blazes.

According to Woodwell Climate Senior Scientist Dr. Jennifer Francis, the phenomenon of winds pushing smoke down to the northeastern U.S. has been linked to rapid Arctic warming caused by climate change.

In the upper atmosphere, a fast wind current called the jet stream flows from west to east in undulating waves, caused by the interaction of air masses with different temperatures and pressures, particularly between the Arctic and temperate latitudes.

As global temperatures have risen, the Arctic has warmed two to four times faster than the average global rate. Dr. Francis stated in an interview in the Boston Globe that the lessening of the temperature differences between the middle latitudes and the Arctic has slowed down the jet stream, which results in a more frequent occurrence of a wavy path.
Another factor contributing to the widespread smoke is an ongoing oceanic heat wave in the North Pacific Ocean. The blob of much-above-normal sea water tends to create a northward bulge in the jet stream, which creates a pattern that sends cooler air down to California and warm air northward into central Canada—resulting in the persistent heat wave there in recent weeks. Farther east, the jet stream then bends southward and brings the wildfire smoke down to the Northeast.

“Big waves in the jet stream tend to hang around a long time, and so the weather that they create is going to be very persistent,” Dr. Francis said. “If you are in the part of the wave in the jet stream that creates heat and drought, then you can expect it to last a long time and raise the risk of wildfire.”

The wildfires are also decimating North American and Canadian boreal forests, the latter of which holds 12 percent of the “world’s land-based carbon reserves,” according to the Audubon Society<./a> And three quarters of Canada’s woodlands and forests are in the boreal zone according to the Canadian government.

“The surface vegetation and the soil can dry out pretty dramatically given the right weather conditions. For this fuel, as we call it in fire science, it often just takes one single ignition source to generate a large wildfire,” said Woodwell Climate Associate Scientist Dr. Brendan Rogers.

As the climate continues to warm, Dr. Rogers said the weather conditions that lead to fuel drying and out-of-control wildfires also increase. This creates a feedback loop. Heat waves caused by greenhouse gas emissions increase the prevalence of wildfires. The fires in turn destroy these natural carbon sinks and, in turn, speed up climate change.

While the ultimate solution to breaking this feedback loop lies in reducing emissions and curbing climate change, Dr. Rogers and other researchers at Woodwell Climate have conducted research into fire suppression strategies that could help prevent large boreal fires from spreading and help keep carbon in the ground.

A study conducted in collaboration with Woodwell and other institutions found that suppressing fires early may be a cost-effective way to carbon mitigation. Woodwell Climate’s efforts also include mapping fires, using geospatial data and models to estimate carbon emissions across large scales, and looking at the interplay between fires and logging.

“Reducing boreal forest fires to near-historic levels and keeping carbon in the ground will require substantial investments. Nevertheless, these funds pale in comparison to the costs countries will face to cope with the growing health consequences exacerbated by worsening air quality and more frequent and intense climate impacts expected if emissions continue to rise unabated. Increased resources, flexibility, and carbon-focused fire management can also ensure wildlife, tourism, jobs, and many other facets of our society can persevere in a warming world,” Dr. Rogers said.

oncoming storm front
A sudden flip in weather conditions—from a long hot and dry period to a parade of storms, for example, or from abnormally mild winter temperatures to extreme cold—can cause major disruptions to human activities, energy supplies, agriculture, and ecosystems. These shifts, dubbed “weather whiplash” events, are challenging to measure and define because of a lack of consistent definition. A new study demonstrates an approach to measuring the frequency of these events based on rapid changes in continent-wide weather regimes.

The study indicates that, while the frequency of whiplash events in recent decades has not changed substantially, future model projections indicate increases will occur as the globe continues to warm under a thicker blanket of greenhouse gasses. In particular, the researchers find whiplash will increase most during times when the Arctic is abnormally warm, and decrease when the Arctic is in a cold regime—something that will occur less often as the planet warms.

Examples of weather whiplash during 2022 so far include a long, hot, drought in western U.S. states during early summer that was broken by record-breaking flash flooding; exceptionally wet and cool conditions during June in the Pacific Northwest replaced by a heat wave in July; a record-warm early winter for most south-central states followed by a cooler-than-average January and February; and a spell of 67 consecutive hot, dry days in Dallas, TX, broken by the heaviest rains in a century.

“The spring and summer of 2022 have been plagued by weather whiplash events,” said lead author, Dr. Jennifer Francis, Senior Scientist at the Woodwell Climate Research Center. “A warming planet increases the likelihood of longer, more intense droughts and heat waves, and we’re also seeing these spells broken suddenly by heavy bouts of precipitation, which are also fueled by the climate crisis. These sudden shifts are highly disruptive to all sorts of human activities and wildlife, and our study indicates they’ll occur more frequently as we continue to burn fossil fuels and clear-cut forests, causing greenhouse gas concentrations to rise further.”

Co-author Judah Cohen, Principal Scientist at Verisk AER noted that these phenomena are tightly linked to regional warming in the Arctic.

“We know the Arctic region is experiencing the most rapid changes in the global climate system. Evidence is growing that these profound changes are contributing to more extreme weather events outside the Arctic, and this influence will only increase in the future,” said Dr. Cohen.

This year in climate, 2021

A look back at a turning point year for climate change

As another year passes with mounting emissions, we take stock of the big moments for climate change in 2021, from extreme weather events to steps forward on policy. Here’s is a look back at a potentially pivotal year for climate change:

Climate change is now front and center due to deadly weather

This year, the clear repercussions of climate change were impossible to ignore. Climate change worsens extreme weather, making Earth’s formerly reliable systems much more unpredictable. Widespread extreme weather events in 2021 had deadly consequences for the people caught in their paths.

In February, Texas experienced a cold snap that killed 210 people across the state and left millions without power for several days. The freeze was exacerbated by the stretching of the stratospheric polar vortex—a pool of cold air high over the poles that is usually hemmed in by strong westerly winds. When the polar vortex is disrupted from its typically circular shape, it can cause the jetstream to waver and plunge farther south, which can bring unusually cold temperatures farther south. Research has connected rapid warming and sea-ice loss in the Arctic north of western Russia with more frequent warping of the polar vortex, which could mean more of these extreme events in the northern hemisphere.

In the summer, temperatures swung in the extreme opposite direction as a heat wave settled over the Pacific Northwest, breaking records. Temperatures neared 120 degrees Fahrenheit, melting power cables, buckling brick roads, and causing sudden deaths to spike across the region. This phenomenon was also caused by a wavering jetstream that allowed an intense and persistent high-pressure system to trap the heat over the Northwest.

Hotter than average weather also led to record fires this year. In Siberia, the return of the fire season to the boreal forests brought with it blazes larger than concurrent fires in Greece, Turkey, Italy, the U.S. and Canada combined. Drought in Brazil, paired with rising deforestation rates, led to increased fires in the Amazon.

And finally, just this month the Midwestern U.S. was hit by a devastating string of tornadoes, more severe than usual for this time of year, marking the deadliest December tornado outbreak in history. Tornados are tricky to study, so an understanding of how their prevalence will fluctuate with climate change is not yet clear, but the conditions fostered by a warmer atmosphere are amenable for tornado formation.

Increased awareness has led to increased momentum

The undeniable severity of the climate impacts of 2021 has enforced the urgency of cutting emissions. This year, climate action began building the momentum it should have had two decades ago, with more than half of Americans concerned or alarmed about the issue, and governments and private sector organizations across the globe stepping up their commitments to tackling it.

In November, Glasgow, Scotland hosted the 26th annual COP—hailed by some as our “last, best, hope” for successful international cooperation on Climate Change. Although the larger negotiations were not as ambitious as necessary to confidently limit warming to 1.5 degrees Celsius, several steps still pushed the world forward. The conference opened with a pledge from 141 nations to end deforestation by 2030, accompanied by $19 million from governments and private sector groups—a large portion of which was dedicated to supporting Indigenous groups.

Earlier in the year, the U.S. Securities and Exchange Commission (SEC) made a request for public comments on a potential requirement for companies to disclose climate risk to their investors. With the impacts of climate change becoming more immediate, the demand for greater certainty around personal risk has grown. The SEC’s inclusion of climate risk in its regulations indicates a broader acknowledgement of the need to prepare for the changes to come. The new presidential administration in the U.S. also signaled its intent to address climate in its first 100 days, staffing up with science and policy advisers and calling an Earth Day summit with leaders from 40 nations.

What to watch for in 2022

So what can we expect as we enter another year?

As long as emissions continue at their current pace, so too will warming and its consequences. Storms, fires, and extreme temperature swings will become a more frequent fixture next year and into the future.

On the international stage, next year’s COP will be one to watch. Nations are expected to return with even more ambitious targets than agreed upon this year. The timelines for formal climate action will accelerate. In the U.S. we can expect to see a ruling from the SEC in early next year.

To match demand for more information on climate risk, Woodwell will be spearheading a collaborative climate risk coalition. The goal of the coalition is to produce an annual climate risk assessment for policymakers to aid future decision-making. Woodwell is also continuing its work conducting risk analyses for climate-related heat, flooding, and fire at the municipality level in several new cities. In 2022, Woodwell will be leading the push for more, transparent climate risk analyses.

2022 will also be a year of expanded research into the impacts of climate change, particularly the rapidly warming Arctic. Woodwell projects will expand our understanding of emissions from thawing permafrost and the behavior of Arctic fires, as well as impacts on extreme weather events. Researchers will also be working closely alongside Indigenous communities to both understand how climate change is impacting vulnerable communities, and support them to take part in climate solutions.

‘Summer of extremes’ briefing helps meteorologists connect extreme weather events to climate change

satellite imagery of hurricane Ida
The summer of 2021 has been a summer of extremes. Catastrophic wildfires, drought, flooding and deadly heat waves are all signals of a warming climate, but the nature of that connection is often not well understood by the public. To help deepen understanding of the links between extreme weather and climate change, Woodwell hosted a briefing last week for a group that is always thinking about the weather: meteorologists.

The briefing was led by meteorologist Chris Gloninger of KCCI 8 Des Moines, who moderated a Q&A with Woodwell senior scientist Dr. Jennifer Francis and Assistant Scientist Dr. Zach Zobel. Dr. Francis and Dr. Zobel provided attendees with insight into how weather events, like the recent hurricanes Henri and Ida or flash flooding in Tennessee for example, are exacerbated by climate change.

Meteorologists, tasked with preparing local communities for changes in the weather, are uniquely positioned to communicate the role of climate change in weather patterns to a broad public audience. According to Dr. Francis, meteorologists are “often the only scientists that people come into contact with.” Which means they have a valuable opportunity to shape people’s perception of weather events as a consequence of climate change.

For Dr. Zobel, a meteorologist by training himself, one of the best ways to make those connections is by highlighting the specific elements of extreme events that the science shows are clearly linked to warming.

“Rather than focus on the storm itself, focusing on the features within the storm that climate models and observations show are clearly going to increase,” Dr. Zobel said. He cited the all time record for the amount of rainfall in one hour in New York that was broken during Henri. Heavy bursts of precipitation are likely to become more common with climate change, as a warmer atmosphere can hold more water vapor.

Meteorologists joined the briefing from across the United States, and were interested in the best ways to communicate climate science across diverse audiences, some of which might not be familiar with or accepting of climate science. Dr. Francis used the example of farmers in the Midwest who have reported more persistent weather conditions—longer droughts or storms—affecting their crops.

“If we can take that and link it back to how we think climate change is starting to cause more persistent weather patterns, that is something we can talk to them about that is really affecting how they do their business, how they live their lives, and is certainly something they are seeing every day,” Dr. Francis said.

Dr. Zobel also emphasized the need to communicate climate change in terms of personalized, individual impacts.

“Until we are able to do that, climate change may seem like a distant, far away problem,” Dr. Zobel said. “Once people see it’s affecting them locally they tend to re-evaluate.”