Regional/Local Impacts of Global Warming in the Great Lakes

[wdmn]

Great Lakes Ice Cover Nearly a Record Low For Early January

https://weather.com/news/news/2020-01-08-great-lakes-ice-cover-low-early-january

"Ice cover throughout the Great Lakes is nearly a record low for early January because warmer than average temperatures have dominated the region so far this winter.

The total ice coverage on the five Great Lakes was 1.5% on Tuesday, or the second lowest for Jan. 7, according to NOAA's Great Lakes Environmental Research Laboratory (GLERL). Ice cover on that date has averaged about 13.9% since records began in 1973.

...

Great Lakes ice cover usually peaks later in winter, sometimes in late February or early March. The current conditions are not reflective of what the peak might be in the next month or two because that will depend on long-term weather patterns.

Last year started out nearly as slow, but then colder conditions moved in and the ice cover peaked at 80.9% in early March. That's much above the average yearly peak of 55.7%.

On the flipside, early January 2017 had about 3% ice coverage. Persistently mild temperatures through February that year prevented ice growth and the season only had a maximum peak coverage of 19.4%.

What really stands out is Lake Erie has no ice coverage right now.

Lake Erie is the shallowest of the Great Lakes, which usually makes it freeze up fastest. The lake's ice coverage has averaged 27.7% by Jan. 7.

The lack of ice on Lake Erie could keep it primed for lake-effect snow deeper into the winter compared to normal unless more persistent cold air arrives soon.

All of the other four Great Lakes are also trailing behind their average ice cover for this time of year."


Attached is a graph showing the date of maximum ice extent on Lake Superior, from 1973 to 2019. Note that 50 days is Feb. 19th and 63 Days is March 4th.

Line is five year running average.

[Shared Humanity]

I live in Chicago and beaches line the shoreline. They are being destroyed this year and many will not exist this spring. On my drive to work this morning, there is a gorgeous beach house built in the early 1900's on the south side that has waves rolling through it. All of these old structures are masonry and brick with more roof than walls so this one will survive the storm. The beach is another story.

[Tom_Mazanec]

Chicago More Vulnerable to Climate Change than Miami, Says ‘Death and Life of Great Lakes’ Author
https://news.wttw.com/2020/01/16/chicago-more-vulnerable-climate-change-miami-says-death-and-life-great-lakes-author

At the point when Egan was writing his book, which synthesizes a decade of his reporting on the lakes for the Milwaukee Journal Sentinel, Lake Michigan was at a near record low. In 2020, it’s at a near record high.

“In the past, it would take a quarter-century to go from low to high. We just did it in five years,” Egan said.

Going back to the 1840s, Lake Michigan has reliably peaked or bottomed out within 3 feet of an average level, for a total 6-foot swing. “Chicago was built on that assumption, so was Milwaukee,” said Egan.

The new norm could be 5-foot variations, for a 10-foot swing, he said, and that’s cause for alarm.

[Tom_Mazanec]

My cousin ran a fishing charter on Lake Erie, so I know a little about this:
Voluntary Measures
Lake Erie turns toxic every summer. Officials aren’t cracking down on the source.
https://grist.org/food/ohio-officials-know-how-to-stop-lake-erie-from-turning-toxic-but-no-one-will-do-it/

The algae are natural. This slow-motion crisis, however, is largely manmade — and neither the federal government nor states are effectively cracking down on the major contributor.

[TerryM]

^^
Good link - Thanks!


The scum is heavy on our Canadian shores.
It's as bad as what we used to see in Lake Mead Nevada, and the fish there all turned into females at one point!


I wonder if this is somehow related to the lowering testosterone levels recorded in NA males?
Terry

[wdmn]

Great Lakes Ice Cover Nearly a Record Low For Early January

https://weather.com/news/news/2020-01-08-great-lakes-ice-cover-low-early-january

"Ice cover throughout the Great Lakes is nearly a record low for early January because warmer than average temperatures have dominated the region so far this winter.

The total ice coverage on the five Great Lakes was 1.5% on Tuesday, or the second lowest for Jan. 7, according to NOAA's Great Lakes Environmental Research Laboratory (GLERL). Ice cover on that date has averaged about 13.9% since records began in 1973.

....

What really stands out is Lake Erie has no ice coverage right now.

Lake Erie is the shallowest of the Great Lakes, which usually makes it freeze up fastest. The lake's ice coverage has averaged 27.7% by Jan. 7."

UPDATE: as of February 5th the ice cover for the Great Lakes is at 6.5%, meaning it is at half the average for a month ago.

Lake Erie continues to be ice free.

[be cause]

I was wondering why bbr had been so quiet this winter . I just looked .. his last post was on my dad's 100th birthday .. his last visit to the forum was on my 60th ..

 certainly this winter thus far has run very differently to his expectations .. and encouraged hibernation ..   ..  b.c.

[blumenkraft]

No, he's banned.

He said some very very stupid shit and Neven had enough.

[dnem]

Yes, I summed up this winter's strong PV, cold arctic/warm continents, low snow cover weather as the "bbr is gonna have to wait winter" over on the freezing thread a few days ago!  Seriously, I am seeing more references in more places to the general idea that despite decent SIE gains this winter, we are set up for an early and fast warmup this spring.  It will be fascinating to see if this verifies.

[wdmn]

No, he's banned.

He said some very very stupid shit and Neven had enough.

It's too bad it had to be that way. He was a lot of fun at times.

[gerontocrat]

& bbr would have not enjoyed the season so far.

Ice cover
5 Feb 2020  6.5%
5 Feb 2019  36.9%
5 Feb 2018  50.3%

Also when the ice cover is low, one expects to hear about "Lake-Effect Snow", but this season very few instances reported on Wunderground.news.

All part of low snow at low latitudes, high snow at high latitudes?

Local businesses complaining - no ice fishing

https://www.glerl.noaa.gov/res/glcfs/compare_years/

Water Levels
https://weather.com/news/weather/news/2020-02-04-great-lakes-record-high-mean-average-level-january

Water levels on the Great Lakes were much higher than average in January, setting new high-water records on three of those lakes. The mean water level for the month, calculated by taking the average of each day's water level, topped the all-time January record on lakes Huron, Michigan and Superior, according to the U.S. Army Corps of Engineers.

January's mean water level on lakes Huron and Michigan was 581.56 feet, surpassing the previous record for January of 581.3 feet set in 1987 (the two lakes are treated as the same lake by the Army Corps of Engineers). The long-term average for January is 578.41 feet.

Lake Superior, the largest and deepest of the five Great Lakes, also broke its all-time January record-high water level. The mean water level last month was 602.72 feet, just above the previous record of 602.69 feet set in 1986. January's long-term average is 601.48 feet.

The two remaining Great Lakes, Erie and Ontario, did not break any records in January, but their water levels were still significantly above average for the month.

Lake Erie had a mean water level of 573.49 feet last month, narrowly missing its all-time January record high by 0.2 feet (2.4 inches). That's more than 2 feet higher than average for the month.

Lake Ontario's mean January water level was only about 0.35 feet shy of its all-time record for that month, or about 1.6 feet (19 inches) higher than the long-term January average.

Water levels across all of the Great Lakes on Jan. 31 ranged from 4 to 27 inches higher than the levels on the same date last year, the Army Corps of Engineers said in its latest weekly report.

Levels on lakes Huron, Michigan and Superior are expected decline over the next month, falling 1 to 3 inches by early March. Still, water levels will remain significantly higher than average heading into the spring. Lake Erie's level is predicted to hold steady, while Lake Ontario is forecast to rise an inch between now and early March.

"High water levels and potentially record-high water levels are expected to persist for at least the next six months, so flood-prone areas are expected to remain vulnerable," said the Army Corps of Engineers.

Why Are Water Levels So High?
The Great Lakes set several monthly records for highest levels from May through August in 2019, and lakes Erie and Ontario set all-time records last summer.

The reason for the persistently high levels in the Great Lakes? Excessive precipitation in the region.

Above-average precipitation has plagued the Midwest for most of the past year. The dominant pattern has brought a parade of storms that dumped heavy snow and rain in the central U.S. since late last winter. Most locations in the Midwest had one of the 10 wettest years on record in 2019.

[blumenkraft]

It's too bad it had to be that way. He was a lot of fun at times.

Well, i don't know. Remember, when he blamed the french revolution for climate change. That's just a tick too insane.

BTW, only recently i learned that Joseph Goebbels had a similar talking point about jews and the french revolution. Verry crazy Nazi stuff...

[The Walrus]

& bbr would have not enjoyed the season so far.

Ice cover
5 Feb 2020  6.5%
5 Feb 2019  36.9%
5 Feb 2018  50.3%

Also when the ice cover is low, one expects to hear about "Lake-Effect Snow", but this season very few instances reported on Wunderground.news.

All part of low snow at low latitudes, high snow at high latitudes?

Local businesses complaining - no ice fishing

https://www.glerl.noaa.gov/res/glcfs/compare_years/

Water Levels
https://weather.com/news/weather/news/2020-02-04-great-lakes-record-high-mean-average-level-january

Water levels on the Great Lakes were much higher than average in January, setting new high-water records on three of those lakes. The mean water level for the month, calculated by taking the average of each day's water level, topped the all-time January record on lakes Huron, Michigan and Superior, according to the U.S. Army Corps of Engineers.

January's mean water level on lakes Huron and Michigan was 581.56 feet, surpassing the previous record for January of 581.3 feet set in 1987 (the two lakes are treated as the same lake by the Army Corps of Engineers). The long-term average for January is 578.41 feet.

Lake Superior, the largest and deepest of the five Great Lakes, also broke its all-time January record-high water level. The mean water level last month was 602.72 feet, just above the previous record of 602.69 feet set in 1986. January's long-term average is 601.48 feet.

The two remaining Great Lakes, Erie and Ontario, did not break any records in January, but their water levels were still significantly above average for the month.

Lake Erie had a mean water level of 573.49 feet last month, narrowly missing its all-time January record high by 0.2 feet (2.4 inches). That's more than 2 feet higher than average for the month.

Lake Ontario's mean January water level was only about 0.35 feet shy of its all-time record for that month, or about 1.6 feet (19 inches) higher than the long-term January average.

Water levels across all of the Great Lakes on Jan. 31 ranged from 4 to 27 inches higher than the levels on the same date last year, the Army Corps of Engineers said in its latest weekly report.

Levels on lakes Huron, Michigan and Superior are expected decline over the next month, falling 1 to 3 inches by early March. Still, water levels will remain significantly higher than average heading into the spring. Lake Erie's level is predicted to hold steady, while Lake Ontario is forecast to rise an inch between now and early March.

"High water levels and potentially record-high water levels are expected to persist for at least the next six months, so flood-prone areas are expected to remain vulnerable," said the Army Corps of Engineers.

Why Are Water Levels So High?
The Great Lakes set several monthly records for highest levels from May through August in 2019, and lakes Erie and Ontario set all-time records last summer.

The reason for the persistently high levels in the Great Lakes? Excessive precipitation in the region.

Above-average precipitation has plagued the Midwest for most of the past year. The dominant pattern has brought a parade of storms that dumped heavy snow and rain in the central U.S. since late last winter. Most locations in the Midwest had one of the 10 wettest years on record in 2019.

The reason that many areas have not had lake effect snow piling up is that the temperatures are too warm.  Most of the precipitation is falling as rain.  Many areas saw one of their rainiest on record.  Further north, areas in upper Michigan have exceeded average snowfall to date by 1-3 feet.  Similarly, northern Minnesota is having one of its snowiest winters in record.

[Shared Humanity]

No, he's banned.

He said some very very stupid shit and Neven had enough.

It's too bad it had to be that way. He was a lot of fun at times.

In an early November record breaking cold snap in the U.S., he stated that this was clear evidence of the beginning of a regional ice age that would become more generalized over time.

[Shared Humanity]

Chicago is know for its Lake Michigan shoreline, a continuous stretch of parks and beaches. They are taking a beating. Entire beaches have disappeared this winter.

[wdmn]

The average time for peak ice on the Great Lakes is here, and cover continues to be exceptionally low. Extent for February 24th is the 4th lowest on record (back to 1973), behind 2012 (4.56%), 1998 (6.2%), 2002 (6.4%). It is only the fourth time that extent has been below 10% for this date.

February 24th
2010-2019 avg: 37.34
2018: 50.5%
2019: 57.2%
2020: 9.0%

While Lake Superior tends to peak during the first week of March, and some growth in extent is expected over the next week, Lake Erie's maximum extent is usually reached around the 18th of February. These two lakes currently show the largest divergence from the mean.

[The Walrus]

Looks like we will witness the consequences of open water on the Great Lakes.  Areas downwind are forecast to receive several feet of snow this week.

[Sigmetnow]

20-foot waves may be coming to the Great Lakes due to approaching winter storm

Lake Ontario could get waves of up to 20 feet near its center, Guy said. Waves as tall as 10 feet could form on Lakes Huron and Erie, peaking on their southern shores.
...
Meantime, blizzard warnings are posted in New York, just south of Buffalo and north of Syracuse, where 2 to 3 feet of snow combined with 40- to 50 -mph winds will make for whiteout conditions.

Travel will be nearly impossible, he said.
...
Increased snowfall this season in the region owes in part to low ice coverage on the Great Lakes, as moisture continues to feed precipitation, Guy said. The lakes this week are 9% covered in ice, compared with a typical 42% average coverage, he said. That's the lowest on record since 1973. ...

https://www.cnn.com/2020/02/27/weather/14-foot-waves-great-lakes-winter-weather/index.html

[ZeaLitY]

Lost my entire weekend reading my way through the forums! Big thanks to Tom and everyone else for amassing such a great collection of news and studies here. The collapse subreddit has gotten shorter and shorter on real climate change content over the years, and so it'd be wonderful to call this a new home.

The Great Lakes are a primary destination for a lot of people on r/collapse/ and elsewhere who want to escape climate change's worst effects in the short and mid-term. I wasn't immune to this, and have networked with a few Yoopers myself, surprised at how few degrees of separation I had with the area once I started talking to people. This thread has been fascinating as the lake-level rise totally escaped my radar over the last several months (COVID-19, Australian bushfires, etc. dominated headlines instead, I suppose). It's raised a few points that contrast with conventional knowledge that the Great Lakes are the panacea for those looking to survive:

• Superficially seems like if the lake level rise continues, it may pose a flooding problem for the edges of the UP, along with other coastal communities?
• A lot of shallow talk about permaculture in the area notes that the lake effect snow does not particularly disrupt organic farming, and forms part of a helpful cycle—but it seems like the precip changing primarily to rain definitely would be disruptive
• A lot of the other shallow reliance on the area points to the UP being sandwiched between two of the Great Lakes as the best temperature-moderator possible outside oceanic coastal areas—it seems like that's a little jeopardized too if the region's climate has depended on consistent winter ice cover, and this is all disappearing earlier and earlier in the year

One feels a bit like Burgess Meredith in the Twilight Zone when staring down the magnitude of research on things like this to read through (and deduce the intersections between) when possessing so little time. Too much to process. I would absolutely appreciate any speculation on this stuff; thanks in advance.

[oren]

Welcome, ZeaLitY.

[Tom_Mazanec]

The 09.2020 issue of National Geographic has an excellent article on the impact of the shrinking winter on the Great Lakes area starting on page 74.

[The Walrus]

The 09.2020 issue of National Geographic has an excellent article on the impact of the shrinking winter on the Great Lakes area starting on page 74.

According to the graph on page 85, the maximum ice coverage on the Great Lakes has  increased 10% over the past 25 years.  Indeed, the maximum ice coverage reached a maximum in the late 70s, dropped until ~2005, and has been rising since. 

[kassy]

So there is plenty of everything for all in the hardcopy, thanks for the heads up. 

[wdmn]

The 09.2020 issue of National Geographic has an excellent article on the impact of the shrinking winter on the Great Lakes area starting on page 74.

According to the graph on page 85, the maximum ice coverage on the Great Lakes has  increased 10% over the past 25 years.  Indeed, the maximum ice coverage reached a maximum in the late 70s, dropped until ~2005, and has been rising since.

This is slightly misleading. There was a deep trough (in terms of max cover) from 1998-2003, but the 2003-2019 trend for max cover is much lower than the period between 1973-1998 (lower minima, and less frequent maxima).

[The Walrus]

The 09.2020 issue of National Geographic has an excellent article on the impact of the shrinking winter on the Great Lakes area starting on page 74.

According to the graph on page 85, the maximum ice coverage on the Great Lakes has  increased 10% over the past 25 years.  Indeed, the maximum ice coverage reached a maximum in the late 70s, dropped until ~2005, and has been rising since.

This is slightly misleading. There was a deep trough (in terms of max cover) from 1998-2003, but the 2003-2019 trend for max cover is much lower than the period between 1973-1998 (lower minima, and less frequent maxima).

Not really.  We are saying the same thing in different words,  Saying the max cover reached a maximum in the late 70s is more precise, but not much different than stating that it was higher in the period 1973-1998.  Similar with saying that the max ice cover dropped or that there was a deep trough. 

Plotting each individual year results in significant scatter in the data.  Using a 5-year average shows the trends much better.

[wdmn]

The reason it is not more precise to say that it reached a maximum in the 70s is because the data set starts in the 70s. Wording it that way makes it sound like there's a cycle of increasing and decreasing max ice cover (and we are currently increasing again), when we in fact know the long term trend has been one of decline.

Also, 2019 was a low max extent year, which your graph doesn't capture (called 2020 in the below annual graph).

In addition, average annual ice cover (AAIC; amount of ice on the lakes each day during freezing season averaged over the whole season) has been lower during recent maxima than at times of comparably high maxima pre-1998; that is to say, the maximum extent lasts less time. Other than 2014-2015 the freezing seasons are getting shorter and less severe.

[The Walrus]

The reason it is not more precise to say that it reached a maximum in the 70s is because the data set starts in the 70s. Wording it that way makes it sound like there's a cycle of increasing and decreasing max ice cover (and we are currently increasing again), when we in fact know the long term trend has been one of decline.

Also, 2019 was a low max extent year, which your graph doesn't capture.

In addition, average annual ice cover (AAIC; amount of ice on the lakes each day during freezing season averaged over the whole season) has been lower during recent maxima than at times of comparably high maxima pre-1998; that is to say, the maximum extent lasts less time. Other than 2014-2015 the freezing seasons are getting shorter and less severe.

Actually, 2020 was the low max year, and is captured in that graph - it is part of the 5-years surrounding 2018. 

Without data from before 1975, we cannot make scientific claims about the maximum ice coverage prior to then.  Again, the max ice is less and the freezing season is shorter than pre-1998.  Since then, the max ice is more and the freezing season is longer. 

How can we be certain that the long term trend is one of decline, if the past two decades is one of increase?

[Shared Humanity]

Pretty clear from that chart that Great Lakes ice cover is in a long steady decline.

[The Walrus]

Pretty clear from that chart that Great Lakes ice cover is in a long steady decline.

What chart?

[Sebastian Jones]

Pretty clear from that chart that Great Lakes ice cover is in a long steady decline.

What chart?

The chart in post 75 by Wdmn.

[Shared Humanity]

Pretty clear from that chart that Great Lakes ice cover is in a long steady decline.

What chart?

The chart in post 75 by Wdmn.

yes.

[The Walrus]

Pretty clear from that chart that Great Lakes ice cover is in a long steady decline.

What chart?

The chart in post 75 by Wdmn.

yes.

Not sure what you are seeing in that chart, but it clearly does not show either a long or steady decline.

[wdmn]

The reason it is not more precise to say that it reached a maximum in the 70s is because the data set starts in the 70s. Wording it that way makes it sound like there's a cycle of increasing and decreasing max ice cover (and we are currently increasing again), when we in fact know the long term trend has been one of decline.

Also, 2019 was a low max extent year, which your graph doesn't capture.

In addition, average annual ice cover (AAIC; amount of ice on the lakes each day during freezing season averaged over the whole season) has been lower during recent maxima than at times of comparably high maxima pre-1998; that is to say, the maximum extent lasts less time. Other than 2014-2015 the freezing seasons are getting shorter and less severe.

Without data from before 1975, we cannot make scientific claims about the maximum ice coverage prior to then.  Again, the max ice is less and the freezing season is shorter than pre-1998.  Since then, the max ice is more and the freezing season is longer. 

How can we be certain that the long term trend is one of decline, if the past two decades is one of increase?

There are a few ways to answer your superficially rigorous questioning.

1) Non-scientific measurements/anecdotal evidence: Did you read the Nat Geo article? They refer to shipping records which talk of break up and start of season, which do not give us max ice cover, but provide some information on Great Lakes ice.

2) Related information about the physical system and accepted physics: We know that the climate has been warming, including around the Great Lakes (as the Nat Geo article also states), and so we have good reason -- i.e. temperature records and physics -- to believe that ice cover was not lower prior to the 1970s than it is now. Nor do we have reason to believe that max ice would have been significantly higher in the 70s than the 1900-1970 period.

3) Meaning of "long term trend" in the satellite records: Finally, in lieu of other satellite data the long term trend is from 1973-2020. And the trend is one of decline in maximum cover.

It is certainly interesting that there have been some very harsh winters in the 2010s, but without some contextualization your graph and the wording of your post are misleading, as I stated in my previous post.

[The Walrus]

1)  Yes, I did read the article.  The reference to the shipping records was rather vague and did not provide much information.

2)  Like most everywhere, the temperature in Michigan has risen, and recent temperatures are the highest recorded.  However, the average temperature during the 25-year period from 1955-1980 was lower than the previous 25-year period, so the ice cover may have been lower during the 1930-1955 time frame.  Prior to that it was likely much higher. 

3)  Overall, the maximum ice cover is less at the end of the dataset (2020) than the beginning (1973).  A simple linear line will show a declining trend, with an R2 of 0.08.  Even that is highest skewed by the first decade of data.  Remove the first 10 years of data, and there is no trend over the past 38 years.

That does not make my post misleading.  Rather, it is misleading to state that the maximum ice cover is declining, when that decline stopped decades ago.  Since we agree that there was a low or "trough" from 1998-2002, and it had affected the recent trend, to gloss over that by saying that the ice is still in decline is a more misleading statement.

[wdmn]

1) You are right that it was a minor reference, and I wish the article had been more robust; however, what is stated is that, "overall [the shipping records show] a clear signal of human-caused warming since the industrial revolution." Which, as you agree in your second point, would mean also a sign of declining ice cover.

2) Yes, this is true (I will have more on the topic soon). There is no disagreement here, and this seems only to reaffirm my point. It would be cherry picking to say this supported your argument that max ice cover "peaked in the 1970s," because we know there was a c.1930-1954 warm period in the temperature record. Rather, if we're comfortable making that claim, we should be comfortable to say that it peaked in the pre-industrial, early industrial period; i.e. the long term trend -- even outside of the satellite record -- is one of decline.

3) This seems like more cherry picking given what you've already admitted. But, the article also talks about certain lakes having less clear trends, due to shallow waters, etc. If we take, for example, Lake Superior (see below), the last 38 years do indeed show a clear trend, even if the R value is low due to lots of inter-annual variability.

You've done almost the exact same thing as those who were claiming there was a "hiatus" in warming, and you seem to concede (as the evidence clearly shows), that the long term trend has been of decline.

As I stated previously, I do think it is remarkable -- and I wish more discussed in an approachable way -- that we've had some extremely cold winters around the Great Lakes in recent years. However, the minima in max extent continue to be "post-1998" level (and, there are many more years reaching below 30% cover than in the period prior), showing that the system has indeed changed. The inter-annual variability has gone up, and the system seems to be oscillating. If you have good reason to believe that when it settles, lake ice will continue to grow, I would really love to hear your case. I tend to believe -- along with most scientists -- that within the next 30 years, we will see lake ice on the Great Lakes mostly disappear.

Now imagine we had not had this exchange and your original post (that I took issue with) was the last and only word. Can you see all that would have been left out, and how that might be misleading to the casual reader... how it might actually even feed a narrative of climate change denial?

[The Walrus]

You've done almost the exact same thing as those who were claiming there was a "hiatus" in warming, and you seem to concede (as the evidence clearly shows), that the long term trend has been of decline.

As I stated previously, I do think it is remarkable -- and I wish more discussed in an approachable way -- that we've had some extremely cold winters around the Great Lakes in recent years. However, the minima in max extent continue to be "post-1998" level (and, there are many more years reaching below 30% cover than in the period prior), showing that the system has indeed changed. The inter-annual variability has gone up, and the system seems to be oscillating. If you have good reason to believe that when it settles, lake ice will continue to grow, I would really love to hear your case. I tend to believe -- along with most scientists -- that within the next 30 years, we will see lake ice on the Great Lakes mostly disappear.

Now imagine we had not had this exchange and your original post (that I took issue with) was the last and only word. Can you see all that would have been left out, and how that might be misleading to the casual reader... how it might actually even feed a narrative of climate change denial?

I think we can agree on the first three points, except to say that I did not claim that the max ice cover "peaked in the 1970s," just that that was the highest in the given dataset.  Based on the temperature data back to the late 19th century, I said that the max ice cover was, "likely much higher."

To the casual reader that may be the case, but I am less interested in what the naïve may garner from a statement than ensuring that the statement is correct and that deniers are not using it to claim that we are misleading the people with questionable claims.  The wild swings are intriguing, and I do not have an answer as to why recent max ice cover has increased.  It may just be similar to the "hiatus."  If I were to bet, I would bet that future lake ice will be less.  However based on recent winters, I would not wager too much.  I do not believe that the lake ice will disappear within he next 30 years.  We may experience lower maxima, and possibly lower minima, but the temperature rise is unlikely to prevent the lakes from freezing.  Very difficult to prevent the water from freezing when the air and land temperatures are well below zero.

[wdmn]

Some related data, which further confirms the overall warming trend's impact on ice around the Great Lakes, can be found in the table below (taken from the paper Historical Trends in Lake and River Ice Cover in the Northern Hemisphere, Magnuson et al, 2000).

Table shows statistically significant changes in freeze, and break-up date for many areas around the Great Lakes (look for Michigan, Minnesota, NY, Wisconsin or Ontario).

[wdmn]

The following is the first of two posts, due to the large number of images.


Satellite records for Great Lakes ice cover only go back to the winter of 1972-1973 (referred to as 1973).

There are two annual measurements for each of the five Great Lakes plus Lake St. Clair for this time period. One is the maximum ice extent reached during the season in percentage of lake cover, and the other is average annual ice cover (AAIC), which is calculated by averaging the extent recorded on each day of the freezing season.

Living near Lake Superior, I took some interest in both temperature records for the area, and the ice data for the late.

Looking at the data, I found that extreme minimum temperatures between Dec. 1 and March 31st from some weather stations (Sault Ste Marie Canada (-20C or lower), Sault Ste Marie, Michigan (-17.78C or lower), and Duluth, Minnesota (-20C or lower)) around Lake Superior gave a very strong correlation with AAIC (0.899085 SSM Canada; 0.8670059 Duluth; 0.9098268 SSM Michigan), and that summing these figures across stations produced an even stronger correlation (Fig 1, correlation = 0.9287145). (Unfortunately, due to incomplete data sets I could not look at other weather stations around the lake).

An ad-hoc mathematical relationship was found between the count of these extreme cold days and AAIC for each weather station, and was again strongest when the counts for all three were summed (Fig. 2, r2 = 0.8927). This relationship allowed for some predictive power during the freezing season (by using two week forecasts, for example), which was of use due to an interest in having forewarning of the possibility of ice bridges forming to some of the large off-shore islands on the lake. The latter being possible because AAIC has a high predictive power for max ice extent (Fig 3, r2 = 0.9358), and max ice extent is a good predictor of years when ice bridges formed.

The temperature data for Duluth went back as far as 1949, SSM Canada to 1962, and SSM Michigan to 1932. Because of strong relationships between extreme cold days and AAIC, and AAIC and max extent, as well as the the count of days reaching -17.78C or below in SSM Michigan, and the count of days reaching -20C or bellow in SSM Canada (Fig 4), I have been able to create some back projections of both AAIC and max extent extending to 1932, which I will share in the following post.

[wdmn]

This is the second of two posts.

Fig 5 Shows Annual Average Ice Cover (AAIC) for Lake Superior from 1973-2020 (Yellow) and several back projections of AAIC.

The first back projection (Green: 1962-1972) was made using the actual data from all three weather stations, since all three went back to at least 1962. R2 for model = 0.8927.

The second back projection (Red: 1949-1961) was made using the actual data from Duluth and SSM, Michigan. R2 = 0.8676

The third back projection (Orange: 1949-1961) was made using the same model used for the first (green) projection, by combining the data from Duluth and SSM, Michigan, with a projected count of extreme days in SSM, Canada (using the relationship shown in Fig 4, above; R2 = 0.8842).

Finally, a back projection (Blue) was made fro the period 1932-1972 using the actual count from SSM, Michigan, combined with the projected count for SSM, Canada. This was done through 1972, so that the period of 1962-1972 could be compared between the strongest model (Green) and the one with the most extrapolation (Blue).

Fig. 6 Shows a the maximum extent of ice cover on Lake Superior (Yellow: 1973-2020), and several back projections of maximum extent from 1932-1972 based on the projections of AAIC outlined above. The relationship between AAIC and max extent is given in Fig. 3, above.

The colours are the same as those on Fig 5.

The data suggests that while Lake Superior experienced somewhat lower ice levels from the late 30s until the late 50s, compared to the period from 1959-1980 (a pattern reflective of the global mean surface temperature anomaly; Fig 7), the general trend has been of a decline in both area covered, and maximum extent. With the exception of 1932 (known as the year without a winter), a notable decline in minima for both AAIC and max extent can be observed beginning in the 1980s, and particularly after the 1998 "super" el nino. Also of note is the increased inter annual variability, especially notable in the maximum extent data over this same period.

Whether these models have any value, or could be improved upon is an open question, but I found them to be interesting enough to be worth sharing. I hope that some of the readers of this forum agree.

[oren]

I hope that some of the readers of this forum agree.

Indeed.

[The Walrus]

I hope that some of the readers of this forum agree.

Indeed.

Same here.  Your graphs seem to add evidence to the 70s (and perhaps 60s) having higher ice cover than the preceding and ensuing years.  Yes, the general trend is a decrease from those higher years.  After 2012, the trend seems more tenuous.

[wdmn]

Thank you Oren and The Walrus,

@The Walrus
Yes, the graphs show a peak in the period you had pointed out. During our earlier conversation I expected ice levels to be slightly lower prior to that based on the temperature data I had already processed, and the global surface temperature record.

However, I have since added another data set to my modelling (from Houghton, Michigan). The data set was missing 4 years, but still produced a strong predictive relationship to AAIC for temperatures reaching -17.78C (0F) or below when this count was summed with other stations.

So, I was able to model the years 1962-1972 with this new relationship (r2 = 0.8909; Black on Fig 8 below), and the years 1950-1961 with a relationship between Duluth, SSMUS and Houghton (r2 = 0.8697; Purple on fig. 8 below). Fig 8 (the first one below) shows these new models along with the ones provided in the previous graph of AAIC (Fig 5 above).

The average of these models (Fig 9 below) shows a less dramatic change between the late 30s until the late 50s and the subsequent period. This is also reflected in the average of the models for maximum extent (Fig 10 below).

As for the post 2012 period, it is so short, and includes what is clearly an extremely exeptional year in 2014 (compare the AAIC for this year with the rest of the data set to see how exceptional it is), that I am hesitant to say that the "pause" in the trend is anything more than short term noise. Unfortunately, only time will tell.

[The Walrus]

Very nice wdmm. 
We two seem to be the most interested in this post due to living within the Great Lakes.