photo by Kelley Elliott, courtesy the NOAA Photo Library
Toward the end of last week, a determined polar bear would have had to paddle hundreds of miles from Alaska’s northwestern coast to hunt for seals at the edge of the polar ice pack.
At the same time, a fast-moving snowshoer (we are talking DC Comics fast, here) could have made it from Oregon to Ontario via Denver without changing footwear, except for a quick scramble down the Snake River Canyon. So, for this time of year, there was very little ice in the Arctic Ocean, but a ridiculous amount of snow in the western U.S. and Canada.
What’s happening is interesting. What isn’t happening – at least not lately – may be even more so.
It seems that for more than a decade, the size of the polar ice cap has been more or less frozen rather than shrinking, albeit frozen at a historically small summertime level. The resulting expanses of open water at the top of the world are influencing weather to the south in complex and sometimes unexpected ways.
The folks at the National Snow and Ice Data Center, who keep track of the area of Arctic sea ice via satellite, rank 2019 as tied with 2007 and 2016 for second-lowest summer ice minimum in the satellite records, which go back to 1979. After tracking at record-low levels since early in the year, the rate of decline in ice extent slowed late in the summer, which is why this year’s minimum did not come close to reaching the satellite-era low that was recorded in 2012.
Still, the 13 lowest sea-ice minima in the satellite record have all been in the past 13 years. This is pretty much what climate scientists and those who cite their work have forecast since around the start of this century. What they did not predict was that the decline of Arctic summer sea ice would, in effect, pause for a decade or more. Only lately are some mainstream climate watchers beginning to note that since 2007, there has been no clear trend in summer sea ice at all. We have simply seen some fairly wide fluctuations around a historically low mean.
In contrast, the 13 years immediately preceding 2007 represented the fastest rate of decline observed by the satellites. So it would seem the Arctic ice pack fell off a cliff and then hit bottom – at least for now.
This has also been true of sea ice volume, which is a function of both the area and thickness of the ice. Since falling to a modern low in 2007, the volume of ice has more or less fluctuated according to how much ice melts in the polar summer or is flushed out of the Arctic into warmer seas by currents and winds.
In 2007 and 2009, former Vice President Al Gore declared that the Arctic sea ice might be completely gone by the middle of this decade. He was talking about summer ice, of course, but even granting him that much license, his claims were alarmist and misleading. A 2013 review of various climate models, cited by Snopes in a 2017 report on Gore’s claims (Snopes assigned subsequent reports of Gore’s claims a “mixed” rating for truthfulness), forecasted ice-free Arctic summers starting sometime between 2020 and 2040.
It is now clear that the summer of 2020 will not bring an ice-free Arctic, or anything close. A polar bear would bet his last seal on it. It would take another fall off a climatological cliff to make summer ice disappear by 2030. We won’t know about 2040 until we get closer to that date.
By this point, however, we ought to at least acknowledge a few things. Climate has definitely changed, and human activity is a factor in that change; readily measurable changes in the atmosphere’s chemical composition and its resulting physical behavior make this self-evident. At the same time, specific forecasts about future climate behavior are less certain. The more specific and longer-range the forecast, the larger the margin of potential error.
The climate in any particular location, as well as the planet as a whole, is the product of a complex interaction of sunshine, heat absorption and radiation (which is what is mainly affected by human activity), cloud cover, ocean temperatures and currents, and many other variables. Even after more than 50 years of computerized weather forecasting, accuracy is quite poor beyond a time frame of a few weeks to a few months. Climate forecasting at time horizons of decades is apt to be even poorer, especially specific forecasts about a particular place, such as whether California will be wetter or drier in the future than in the past, or about a specific effect, such as whether the oceans will drown our coastal cities in our grandchildren’s lifetimes.
The most recent estimates of global sea level rise are in the neighborhood of 3.4 millimeters per year. That’s about one inch every 7.5 years, or one foot in 90 years. Could things get worse at a worsening rate? Yes. But that is only one possibility among many, and not necessarily the most likely. Changing sea levels are not just a function of how much ice melts at the bottom of glaciers. They are also a function of how much rain falls on land, how quickly it drains into the sea, how much water is sequestered in the warming atmosphere, and how much snow falls on the still-large glaciers of Antarctica and Greenland. A warmer, moister world could readily lend itself to more snowfall on those glaciers, not less, and a concurrent increase – not decrease – in total water content. That has already happened in Greenland in some recent years, although seemingly not this year (final estimates won’t be available for some time). Our ability to measure and model Antarctic snowfall in particular is limited.
None of this is to dismiss climate as a concern. But if we are going to consider forecasts of future change as a driver of policy, we ought to consider the range of possible outcomes of those forecasts – and start with the degree to which past forecasts have, or have not, been borne out. If we want fact-based decisions, we should start with the observed and observable facts. One of those facts is that the much-discussed Arctic sea ice has not changed appreciably in more than a dozen years.