MIT Climate

If humanity fails to address our runaway climate pollution, we may see 3° C or more of global warming by the end of this century—temperatures which we know in our planet's past have led to 30 or 40 feet of sea level rise and huge changes to rainfall and plant life across the globe. But... *how* do we know that? How can scientists take the temperature of the Earth millions of years in the past? Prof. David McGee of the MIT Department of Earth, Atmospheric and Planetary Sciences joins our Ask MIT Climate series to explain the science of "paleoclimate," which has developed extraordinary tools to do just that. “Paleoclimate is the art of the possible,” he says. “There are only so many things that are left over from thousands of years ago, and even fewer left over from millions of years ago. We have to rely upon natural archives, things that grow or are deposited and somehow record information about the climate around them as they form.” And yet, scientists like McGee have learned to find telltale clues to rainfall patterns in ancient stalagmites, to the makeup of our atmosphere in seafloor sediments, and to past temperatures in layers of Antarctic ice. Combining these pieces of evidence produces a remarkably rich and consistent picture of the Earth's climate going back 100 million years. And while that record tells us that the Earth as a whole is remarkably resilient to big changes in climate, it also shows that the adjustment periods—such as humans are creating right now by burning massive amounts of fossil fuels—can be harsh and dangerous ones.

Have you ever heard a weathercaster explain that climate change made a heatwave "three times more likely," or a storm drop "20% more rain"? On the newest episode of our TILclimate podcast, Dr. Andrew Pershing of Climate Central, Inc. joins us to talk about the emerging science of "climate change attribution," which increasingly lets scientists draw a straight line from our warming planet to our escalating extreme weather. "When I was in grad school, we were told you can’t connect any one weather event to climate change," Dr. Pershing explains. Changing that took new scientific tools, advances in climate modeling—and a boost from our changing climate itself, which more and more throws weather events at us that simply could not have happened if human actions were not warming the Earth. As our climate evolves, our responses will need to evolve too, and professionals from meteorologists to insurers, from policymakers to the Red Cross, are now drawing from climate change attribution to understand which weather disasters are likely to hit us again and again.

A reader asked us: which industries contribute most to climate change? And although we have very good data on the climate-warming emissions from all kinds of industrial processes, this is still a surprisingly tricky question to answer! For the latest edition of Ask MIT Climate, Sergey Paltsev, Deputy Director of the MIT Center for Sustainability Science and Strategy and senior research scientist at the MIT Energy Initiative, helps us understand how different industries' climate impacts are tallied up. Because while there are certainly industries that do as much to warm the climate as major countries—like the chemicals industry (responsible for 5% of global emissions), steel (7%), and concrete (8%)—once you start looking at the entire pie chart of world emissions, things get a lot more complicated. “In general, you cannot add these numbers up for all industries, because you're going to do double counting,” Paltsev says. When a steel plant ships a load of steel across the sea, the emissions from burning that ship fuel arguably belong to the steel industry *and* the shipping industry *and* the oil industry. So in tackling these emissions, researchers and policymakers can find better information by closely studying?*how*?an industry contributes to climate change, rather than just asking?*how much.* Some emissions are relatively easy to deal with (like from electricity use), while some are much harder (like the chemical reactions that produce climate-warming CO2 when iron turns into steel). Some are strongly in the industry's control (like energy use at a steel plant), some perhaps less so (like what mining equipment is used to get raw iron ore). If all you want is a headline number to call attention to a certain industry, this may sound like picking nits. But in the hunt for solutions, Paltsev says, the details of how our climate pollution enters the atmosphere are far more important than which industries get the blame—so we should learn how to look under the hood of the big, headline numbers.

The Earth has gone through massive climate change before—many times over, in fact!—but human civilization has not. For the latest episode of our TILclimate podcast, we talk to Prof. David McGee of the MIT Department of Earth, Atmospheric and Planetary Sciences: a "paleoclimatologist" who hunts for clues about ancient climates. Prof. McGee joins the show to explain the many changes the Earth has experienced before the 10,000 years of global stability in which complex human societies emerged and grew. Along the way, we'll learn about the scientific tools used to study the distant past, the great cycles of the ice ages, and what it all tells us about the climate change we’re experiencing today. (Like: did you know the next big climate shift we would expect from natural causes—that is, if we *weren't* burning climate-warming fossil fuels—is actually a period of global cooling, sometime in the next several thousand years?) Listen and follow the show to learn more, on Apple Podcasts, Spotify, or wherever you get your podcasts!

We live in a world that's currently dependent on fossil fuels, yet a lot of that dependence is invisible to us. Today, our TILclimate podcast returns for its seventh season to cast a light on the hard work being done to eliminate climate pollution from a place where it's particularly hard to spot: the food we eat. Prof. Jennifer Clapp of the University of Waterloo, IPES-Food and the UN Food Systems Coordination Hub joins the show to walk us through all the ways fossil fuels are used to produce one simple food item: a tortilla chip. Along the way, we'll find out how our farm equipment, transportation systems, barns, mills, packaging plants, fertilizers, herbicides, pesticides and more are contributing to the buildup of heat-trapping gases around our planet—and the alternatives we could and in many cases are putting in place. This special episode is brought to you in collaboration with the university coalition TABLE, whose recent podcast miniseries Fuel to Fork digs deeper into the use of fossil fuels across the global food system. Check out our season 7 premiere at the link below, or follow TILclimate on Apple Podcasts, Spotify, or wherever you listen! https://lnkd.in/gUxsY8vf

Generative AI, with its high energy use, is changing the trajectory of the whole U.S. electric grid—with real implications for climate change. For our Ask MIT Climate series, Vijay Gadepally, senior scientist at the MIT Lincoln Laboratory Supercomputing Center, talks us through the options at hand to navigate the rapid growth of AI while producing as little added climate pollution as possible. Today, the U.S. is quickly building new data centers to handle the massive processing tasks of training AI models and letting users query them once they're released. This has contributed to a situation the U.S. has not experienced for the past 20 years: our energy needs are growing again. And while ideally we would meet this new energy demand with clean, non-climate-polluting sources, in practice it's hard to power round-the-clock data centers with variable wind and solar farms. Until experiments with clean "firm" energy work out—like Google's efforts to power data centers with geothermal energy, or Microsoft's revival of nuclear power—our new data centers are likely to be powered by a buildout of climate-warming natural gas plants, as is already happening in hotspots like Texas and Virginia. Could we create and run AI programs with less energy? Dr. Gadepally believes that developers can get more creative about achieving ever-better AI results without constantly growing their models. “Every time you talk to a generative AI model, you're passing in some data that is flowing through hundreds of billions of parameters,” he says. “And in general, the thinking has been the larger the model—that is, the more parameters it has—the more accurate or realistic content it can produce.” Recent innovations, like the release of the DeepSeek model, show that this is not necessarily the case. Gadepally also says we can make the data centers themselves more energy-efficient. Data centers need cooling, but the traditional method of using air conditioning is already giving away to far more efficient technologies. His lab has also seen impressive results from “power capping,” or limiting the electricity each processor is allowed to draw. “We have seen reductions in energy use for workloads between 15 to 20%," he says. "And it doesn't change the answer in any way. It just makes it a little bit slower, by a couple of percent.” None of these changes on their own will zero out the climate pollution from using AI, but all of them are significant—especially in the context of an energy-hungry type of computing that is being used at a huge scale all around the world. “I like to remind everyone that even small changes are okay,” says Gadepally. “We're doing billions of these things a day. So even a small change, when you multiply it by the number of times we're doing it, can be quite large.” https://lnkd.in/ePvpK2nz

A recent executive order has halted leases and permitting for offshore wind projects in the U.S.—effectively taking any growth in this clean energy resource off the table. For our Ask MIT Climate series, we consider what this means for states like Massachusetts, New York and Virginia, where offshore wind is a cornerstone of future energy plans. Is it a fatal blow, or just a minor setback? It's more the latter, says Joshua Hodge, Executive Director of the MIT - Center for Energy and Environmental Policy Research. “There are many paths to reducing carbon,” Hodge explains. “Removing any one technology, such as offshore wind, still leaves you with other options. Of course, if you remove too many, it gets dicey.” Even within our large energy toolbox, some technologies are more important than others.?Solar energy?and onshore wind (the kind built on land) are the hammers and screwdrivers of clean energy today, useful across the country in projects large and small. If one of these became unavailable, it would be a huge challenge, whether you care about decarbonizing our energy system or just about having cheap, abundant electricity. Offshore wind is more like an electric band saw: a pricey device built for specific situations. The best conditions for offshore wind in the U.S. are found in the Northeast, from North Carolina to Maine. But even here, offshore wind is more expensive, kilowatt for kilowatt, than onshore wind or solar. Its main advantage is that it doesn't compete for space with cities and suburbs, making it attractive in small, dense states like Massachusetts with less open room to build on land. Some states building offshore wind today have many other options for homegrown clean energy. New York has excellent onshore wind resources, and is busily expanding its electric grid to use more upstate wind and even clean hydropower from Canada. Virginia has a large nuclear power fleet and is quickly adding solar. Other states, like Massachusetts, New Jersey and Rhode Island, have less flexibility. But if these states are willing to tackle the challenge of securing out-of-state clean energy resources, and building the transmission lines to transport that power, they too can move forward without offshore wind, remove climate pollution from their energy systems, and lower customers' energy bills at the same time. “If Massachusetts could just build onshore wind in places like Northern Maine and Quebec and upstate New York, where there's a lot of room and fantastic onshore wind resources, that would be much, much cheaper for us,” says Hodge. “There's absolutely no reason why we have to have offshore wind in order to deeply decarbonize the power systems in the Northeast and Mid-Atlantic.” Read more at Ask MIT Climate: https://lnkd.in/e_rREgdV

A reader asks: Is urban tree planting a significant climate solution? It depends what kind of solution you're looking for, explains Prof. John E. Fernández of the MIT Department of Architecture and MIT Environmental Solutions Initiative for our Ask MIT Climate series. Planting trees in cities can be a very valuable form of adaptation, or preparing for the impacts of climate change. But it's a poor, even negligible, form of mitigation—that is, addressing the mounting carbon emissions that are the root cause of that climate change. It's well-known that trees draw carbon out of the atmosphere as they grow. And it's both logical and true that, if we planted enough of them, that could counteract some of the climate-warming carbon that humans are putting into the atmosphere. But, explains Fernández, that would require vast new forests—not the kind of tree-planting we can do in parks and along city streets. “The amount of trees that exist in cities and the amount of new trees that you could plant in cities is trivial in comparison to the number of trees on the planet,” he says. “Even if you did the absolute maximum planting of trees in every city on the globe, it would have a trivial effect on pulling carbon dioxide out of the atmosphere, if any effect at all.” On the other hand, trees do provide a valuable service to city residents. Their shade minimizes the sunlight hitting the ground, streets, buildings and other infrastructure, and they also absorb heat as they take in water from the air and soil. That provides potentially life-saving cooling in cities that are increasingly affected by extreme heatwaves. That is especially important at night, says Fernández. “Climate change is leading to very hot nights, and there are very few ways in which to deal with that,” he says. “One of the ways is to increase the evapotranspiration that comes from plants and trees, especially in the urban space itself.” So while urban tree planting is not effective at dealing with our carbon emissions, it's still an idea that many cities would be well-advised to invest in as the climate changes. https://lnkd.in/diqy9Zh6

What would it take to power our cars and trucks if everyone switched to electric vehicles? Prof. Jessika Trancik of the MIT Institute for Data, Systems, and Society (IDSS) joins our Ask MIT Climate series to help us look at the electricity generation needs of a fully electric vehicle fleet. The first thing to know is that this would be a very large increase in electricity demand. In the U.S., various studies have estimated that an all-electric fleet would make up between 13 and 29% of our *total* electricity use: somewhere in the rough neighborhood of 1,000 extra terawatt-hours of electricity a year, or the equivalent of well over 100 nuclear reactors working around the clock. Luckily, Trancik explains, that doesn’t mean we need vast numbers of new power plants to provide EV charging. What we need to watch is not the total electricity our EVs consume, but their impact on the hours of?peak electricity demand. “What you're trying to avoid here is a scenario where everyone acts in unison to plug in, especially at times when other demands for electricity are peaking,” she says. With smart planning, even a fully electric fleet can mostly get by on the spare capacity of power plants that would otherwise be sitting idle. The key is that EVs, unlike most things we use that consume electricity, can be charged at any time of day. So a smart strategy for vehicle electrification will make it easy and convenient for people to charge their cars when demand for electricity is low, like late at night and in the middle of the workday. Trancik suggests several, mutually supportive ways to meet that goal: installing ample charges at and near workplaces, using delayed charging on home chargers so EVs wait to fill up until demand is low, and charging less for electricity at off-peak hours. “Getting everything to align can be, of course, challenging in the real world,” Trancik adds. “To actually implement this, you need to find ways to incentivize the installation of chargers at the workplace, and also to provide these programs and incentives for that staggered home charging.” https://lnkd.in/d87bPTx5

The transition to clean energy doesn't just demand new power sources like wind, solar, hydro and nuclear. It also requires a big buildout of transmission: the complex network of cables, transformers, substations, switches and communications equipment that brings power from where it's generated to where it's used. In our newest Climate Explainer, energy economist Prof. Paul Joskow of the MIT Department of Economics and MIT - Center for Energy and Environmental Policy Research breaks down why a clean energy system needs more transmission than a climate-polluting one, and the challenges and opportunities of building that transmission. The U.S. Department of Energy estimates that, to use 100% clean energy by 2050, the United States’ current transmission capacity will need to more than double.?This is possible, and would also offer us cheaper electricity. But it demands that we build transmission much faster than we have in the past. Some regions, like Europe, have shared planning organizations that can map out and fund large transmission projects crossing boundaries, making the most of every energy source available. The United States, where some "transmission operators" serve no more than a single state, could benefit from this example. Our energy permitting system is also designed with many, overlapping veto points, with some larger transmission projects needing to seek approval from federal, state and local governments, plus additional agencies responsible for federal and state forests, parks, grazing, and other protected areas. A streamlined permitting process could provide all the same safeguards for landowners and the environment in a fraction of the time, letting us build faster and more reliably. We can also do more with the transmission lines we already have. “Reconductoring” with materials like carbon fiber can let our cables carry much more power than traditional copper and aluminum wire. New monitoring and switching systems can let us safely use power lines at closer to their full capacity, without risking blackouts in worst-case weather conditions. And even new transmission lines can be paired with existing infrastructure, taking advantage of the right-of-way from highways, railroads and canals. Read the full Explainer on the MIT Climate Portal: https://lnkd.in/efRFPiQt