Arctic sea ice hit a historic low this winter, shrinking to the smallest extent ever recorded at its seasonal peak.
NASA and the NSIDC revealed that warmer air, ocean temperatures, and persistent wind patterns have limited both the formation of new sea ice and the survival of older ice. Scientists, using decades of satellite data, warn that this winter’s record low could have dire consequences as the region enters summer with less protective ice than ever before.
Record-Low Arctic Sea Ice in 2025
On March 22, 2025, Arctic sea ice reached its lowest winter peak ever recorded, according to NASA and the National Snow and Ice Data Center (NSIDC). The maximum ice extent measured 14.33 million square kilometers (5.53 million square miles), falling below the previous record low of 14.41 million square kilometers (5.56 million square miles) set in 2017.
Typically, sea ice grows and spreads across the Arctic during the dark, cold winter months. But in recent years, less new ice has been forming, and there’s been a decline in multi-year ice, ice that lasts through more than one season. This winter followed that same long-term downward trend, which scientists have been tracking for decades.
Tracking Ice from Space
To measure sea ice extent, scientists use satellite data to map the surface of the Arctic. They divide the region into a grid and count the total area of all sections that are at least 15% ice-covered. The map above from March 22 shows this year’s extent compared to the median for the same month between 1981 and 2010. That median line marks the middle range of past observations: half of the years had more ice than the line, and half had less.
How Scientists Measure Sea Ice
For their analysis, scientists primarily rely on satellites in the Defense Meteorological Satellite Program (DMSP), which measure Earth’s radiation in the microwave range. The DMSP data are augmented with historical sources, including data collected between 1978 and 1985 with the Nimbus-7 satellite that was jointly operated by NASA and the National Oceanic and Atmospheric Administration (NOAA).
Warm Air, Wind, and Regional Impacts
At the sub-Arctic level, many parts of the region saw sea ice extents well below the 1981-2010 average this winter. For instance, low ice prevailed across the Gulf of St. Lawrence, likely influenced by warm air and ocean surface temperatures, for most of the season. Onshore winds, which can prevent new ice formation, may have also contributed.
A Troubling Outlook for the Summer
“We’re going to come into this next summer season with less ice to begin with,” said Linette Boisvert, an ice scientist at NASA’s Goddard Space Flight Center. “It doesn’t bode well for the future.”
NASA Earth Observatory images by Lauren Dauphin, using data from the National Snow and Ice Data Center. Story by James R. Riordon/NASA’s Earth Science News Team, adapted for Earth Observatory by Kathryn Hansen.
Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.
12 Comments
OK, the second time around (April 1 and April 5) some ‘statisticky’ data was provided. However, the only information actually provided is that March 22, 2025 and March 7, 2017 appear to be more than 2-standard deviations from the 1981-2010 median of the sea ice extent.
It might be the mean; they are unclear because they use the ambiguous term “average” for the graph above; both the median and mean are considered to be types of averages. However, a standard deviation cannot be derived from a median; therefore, the validity of the claimed standard deviation is suspect. It needs explanation. The problem is compounded because when I went to the NASA and NSIDC websites to try to better understand the situation I discovered that they sometimes use different definitions for the sea ice coverage (actual area and an integrated (summed) average that they refer to as an index for the extent, mix standard deviations, interdecile, and interquartile ranges, and most importantly, choose different subsets of the historical data for comparisons. It is therefore, difficult to compare the graphs with no standards for presentation.
As best as I can make out, the 2025 sea ice extent is probably not statistically different from 2015 through 2018, nor are they probably statistically different from each other. That is, any apparent differences may not be real, instead being artifacts of the uncertainty bounds of the measurements. NSIDC remarks on their website “This [March 22, 2025] is the lowest maximum in the 47-year satellite record, with previous low maximums occurring in 2017, 2018, 2016, and 2015.” Furthermore, NASA shows with a graph that March 26, 2024 had a greater ice extent than all other years between 2014 and 2025. Unlike them, I’m not cherry picking, I’m working with the data they provide on their websites.
In summary, it appears to me that there is an attempt to put a spin on the data to raise alarm about a behavior of Arctic ice that has been going on about 20,000-years, with episodic reversals, such as the Little Ice Age.
good try, GW denier. but youre gas lighting obviously
If it is obvious, then even you can explain how and where I’m gas lighting. Point out ANYTHING that I have written that is not true.
BTW, I accept that Earth is warming. What I deny is the influence that humans supposedly have over the amount.
ice kills life
the amount of life on the planet is inversely proportional to the amount of ice on the planet
the only reason the arctic has ice is because we are currently in a major ice age
have been for 2.6 million years
it’s called the Quaternary ice age
typically our eon, the Phanerozoic, is ice free year round even at the poles
it’s extremely cold today
99% of the last 245 million years were warmer than today
We weren’t around then, we’re used to this. If we can warm it up we can cool it down. Just need to figure out how. Fact is the rate of change is unprecedented except in previous mass extinctions, and evidence is were in another one now. Check out honeybees in Kenya and warming.
Completely idiotic. You know nothing of history nor climate.
This time it’s the Artic, three days ago it was the Antarctic : “Antarctica on Thin Ice: Satellite Data Reveals 2025’s Stunning Sea Ice Collapse By NASA Earth Observatory April 2, 2025” ( … Story by James R. Riordon/NASA’s Earth Science News Team)
Both AGW polar-ice catastrophe “Stories” (their description, not mine) put out by the same authors?
What a coincidence – who would have thought, the same doomsday topic peddled, yet again, also by the same organisation.
Thanks again Clyde Spencer in taking the time and using your expertise to point out the lack of scientific rigour shown in both articles.
Thank you!
A somewhat technical article on the subject of an oil company’s study of fossil fuel burning and whether it actually parallels observed climate effects. Make of it what you will.
https://www.science.org/doi/10.1126/science.abk0063
From the article you provided the link for: “Our results show that in private and academic circles since the late 1970s and early 1980s, ExxonMobil predicted global warming correctly and skillfully.”
Just because someone commonly considered an ‘expert’ makes a claim, it doesn’t make it fact. See Figure B for a summary of all the projections from the time period. The earliest one shown isn’t even in the running, having a temperature increase nowhere close to reality. If one is going to propose changes to the energy system of the world, qualitative predictions that vary considerably, and only share a positive slope, are not sufficient. The quantitative predictions should be within at least +/-5% to be considered “skillful,” in my opinion. The first qualitative example is like if your bank sent you a monthly statement that simply said, “You still have funds available,” instead of giving you the details if each and every transaction so that you know how much you have in checking, down to the nearest penny. $0.01 precision in an account with a nominal $5,000 balance is 0.0002%! Yet, people are clamoring to turn the multi-trillion dollar world economy on its head based on estimates that disagree by several tens’ of a percent, even today.
The infamous prediction by James Hansen, which he presented to Congress in 1988, wasn’t as good as a simple linear extrapolation of his own historical data. The ‘business as usual’ scenario (A) was unskillful, it doesn’t come close to what has happened in the intervening 37-years. Scenarios B & C (characterized as “Draconian reductions,” best fit history, but they do so not only for the assumed reductions in anthropogenic CO2 reductions, but also the down-played assumption that there would be two major volcanic eruptions in the near future. Humans have no control over volcanic eruptions, so the same assumption should have been applied to ‘business as usual’; however, it wasn’t!
https://wattsupwiththat.com/2018/06/30/analysis-of-james-hansens-1988-prediction-of-global-temperatures-for-the-last-30-years/
After more than 40-years there is still no general agreement on how much the Global Mean Temperature should increase with a doubling of the amount of CO2 in the atmosphere; the estimates range from less than 1 deg C to over 6 deg C per century. Yet, people like Oreskes are willing to wipe the slate clean and try untested technologies to avoid burning fossil fuels when we don’t have a good grasp of what that would cause in the way of unintended consequences. The extant Global Climate Models (GCMs) are acknowledged by even their modelers as running warm, yet they seem unwilling to ask the critical questions about their numerous assumptions. The GCMs are often characterized as just being physics, and rarely is it acknowledged that there are no computers fast enough to handle the energy exchanges in clouds at the same spatial resolution as the other meteorological parameters. Therefore, they take a short-cut and have cloud-physics experts make their best guess as to the net results of perturbations and use a table lookup substitute for the partial differential equations that are the actual physics; that is called “parameterization.
That’s what I make of the opinion piece called a “study.
The issue for me is not the arguable precision of assorted measurements but the guaranteed precision of human stupidity as resources get used up, wasted etc and that refers to the biosphere we live in. IF anthropogenic global heating (AGH) develops anywhere close to even our most modest predictions, then our stupidity guarantees thatt somewhere between 8.5 billion and 12 billion people due on this planet in the next 100 years are going to have a very rough time of it.
That includes my grandchildren.
Therefore, the precautionary principle comes into play, and that does not include the rich building bunkers. It does, however, include winding down our profligate and unnecessary waste of resources, from water, arable land and hydrocarbon fuels. And that is necessary even if AGH is a total furphy bruited abroad by evil greenies. AGH will just make our pending ocial discomfort occur more rapidly.
Talking about the Precautionary Principle is just arm waving. To apply it, rationally, requires reliable data from a cost-benefit analysis for the speculated event, and a probability threshold for activating your precautions. I believe I previously asked you at what probability of a meteor impact would you start taking action to prevent becoming a casualty, and how much money you would spend to make your actions worth the effort. I didn’t get an answer. Would you borrow as much as you can if it became obvious that it wouldn’t be enough to insure survival?
Inherent in the Precautionary Principle is the assumption that there is a minimal cost solution that would be sufficient to prevent the speculated disaster. There is NO general solution for all potential disasters and no guidelines as to what probability should trigger taking action, nor any guidelines as to how much money is too much for an event of unknown probability. Chauncey Starr demonstrated that people tend to spend money on activities in proportion to the perceived benefit(s). The perceived benefits are usually emotional and heavily influenced by commercial advertising and media propaganda.
I don’t think that you have thought this through! Nor have you examined the unstated assumptions. HCI used to have a slogan that no cost was too high if it saved even one life. Such an emotional position overlooks the fact that focusing on one issue, and devoting all available resources to it, will then result in other problems taking lives. Thus, all that is accomplished is trading lives lost for one reason to a different reason.
Stupidity comes in all sizes and colors — even Kiwi green.