Historic Climates I - The Little Ice Age (1450-1850)

[WYorksWeather]

Hi all,

I thought I'd start this post (which may turn into a series) to have a look at some historic climate changes over time. This is partly inspired by a recent discussion I had in another thread on the Paleocene-Eocene. For each epoch I look at, I'll try to explain what the impacts were regionally and globally, what caused the change (if known) and what lessons can be learned if any.

If this does turn into a series, I'll probably tackle different climate epochs in reverse chronological order. I thought about starting with the 20th century, but that's probably been largely done to death in the historic weather section, so I thought it better to move beyond that. Absent any requests, if I do turn this into a series (mostly dependent on interest), my next post will probably be on the Medieval Warm Period, followed by the Holocene Climate Optimum. Happy to discuss any issues raised in the comments.

What was the Little Ice Age?

In the words of the IPCC (AR6):

Quote

a multi-centennial period of relatively low temperature beginning around the 15th century, with GMST averaging –0.03 [–0.30 to 0.06] °C between 1450 and 1850 relative to 1850–1900.

In English, it's most likely the coldest period of global temperatures in the last 2,000 years, as shown in the PAGES2k climate reconstruction. The time frame that we are interested in is bounded by the black vertical lines I've added to the chart:

As shown, the global impact is relatively modest, and not significantly different from zero.

However, the effects are much more noticeable if you look at smaller geographies, such as the 2010 reconstruction of extratropical Northern Hemisphere temperatures (from 30 degrees N to 90 degrees N) by Ljungqvist below. Temperatures were likely more than half a degree below 1961-1990:

In short, much of the variation is smoothed out on a global scale, but it does seem reasonable to suppose based on available evidence that there were very substantial changes in much of the Northern Hemisphere, at a minimum.

Selected notable climate or climate-related events

• Expansion of glaciers in much of the Alps
• Some of the UK's coldest documented winters (including the legendary 1683-4 winter)
• Swedish army marched across the ice to Denmark in 1658 (The March Across the Belts)
• Widespread crop failures in Iceland
• Abandonment of Norse colonies in Greenland
• Harsh winters in New World colonies

Causes

Causes of the Little Ice Age are still fairly disputed - there are potentially a number of factors which may have acted together. I've listed a few causes, in rough order of how commonly they are cited as a major factor:

• Volcanic activity
• Lower solar activity
• Orbital changes
• Internal climate variability
• Reduced human population

Summary

The Little Ice Age was significantly cooler than the mid to late 20th century across large parts of the Northern Hemisphere, but the impact on a global scale was smaller or possibly even zero. Since this is still relatively modern (compared to climatic variations in the more distant past), we can learn a lot about how societies reacted to the changes. It has also lived on in our cultural memory!

I'll end with an illustration of the Thames Frost Fair in the winter of 1683-84, and one of the famous prints you could have bought on the River Thames (held by the Museum of London, and this one bought by King Charles II!):

 

If you've got this far, thanks for reading.

[Methuselah]

A PBS documentary about the interplay between organic life and minerals:

 

[Roger J Smith]

I am skeptical that reduced human population had anything to do wit the Little Ice Age. Is there any research to document this claim? (I realize there was a reduced population after a century of plague (14th c)). Is it postulated to be related to agricultural practices, otherwise I would be at a loss to imagine any cause and effect in those pre-industrial times (by and large). I would say partly volcanic activity and partly prolonged lower solar activity in two cases, Sporer minimum of 15th century and maunder minimum 1650 to 1710 (also Dalton minimum coincided with another downturn in winter temps). 

I don't think the LIA really ended until around 1910, that is where larger increases began to develop. Definitely 1890 to 1910 is a better end date, the 1880s were a very cold decade in many regions. It could also be argued that the LIA began as early as 1300; that is about where Norse settlements in Greenland began to experience problems. 

A good way to visualize climate of LIA would be our recent coldest and also wettest intervals were about normal (average) in those centuries. There were many severe storms in the North Sea and repeated inundations of coastal areas of Frisian Islands etc. The jet stream was probably at least 3 deg of latitude closer to the equator than its recent positions (in n.h. at least). I recall reading in Lamb's studies that coastal ice often extended from east Greenland to Iceland and that polar bears were able to roam around in Iceland as a result. I don't think sea ice has made it more than halfway across the strait in the past hundred years. 

[WYorksWeather]

1 hour ago, Roger J Smith said:

I am skeptical that reduced human population had anything to do wit the Little Ice Age. Is there any research to document this claim? (I realize there was a reduced population after a century of plague (14th c)). Is it postulated to be related to agricultural practices, otherwise I would be at a loss to imagine any cause and effect in those pre-industrial times (by and large). I would say partly volcanic activity and partly prolonged lower solar activity in two cases, Sporer minimum of 15th century and maunder minimum 1650 to 1710 (also Dalton minimum coincided with another downturn in winter temps). 

I don't think the LIA really ended until around 1910, that is where larger increases began to develop. Definitely 1890 to 1910 is a better end date, the 1880s were a very cold decade in many regions. It could also be argued that the LIA began as early as 1300; that is about where Norse settlements in Greenland began to experience problems. 

A good way to visualize climate of LIA would be our recent coldest and also wettest intervals were about normal (average) in those centuries. There were many severe storms in the North Sea and repeated inundations of coastal areas of Frisian Islands etc. The jet stream was probably at least 3 deg of latitude closer to the equator than its recent positions (in n.h. at least). I recall reading in Lamb's studies that coastal ice often extended from east Greenland to Iceland and that polar bears were able to roam around in Iceland as a result. I don't think sea ice has made it more than halfway across the strait in the past hundred years. 

The postulated reason for cooling would indeed be land use changes - e.g. the human practice of things like clearing land for farming etc. could have already been having smaller impacts over that time period, and the large reduction in the population would have slowed down or even reversed that process (large areas presumably going back to nature). That of course is only a hypothesis, and is very much contested in the scientific community as an explanation. You're right that it is far too soon for any sort of emissions-based warming, and in fact even into the early 20th century this was quite modest relative to today's levels (at most 5-10% of the modern emissions) and would of course have been far lower still in the European late medieval and Renaissance/Enlightenment periods.

In terms of the solar / volcanic balance, there is a good reason for saying that volcanic is a major, or even dominant factor (though of course it's unlikely to be 100% of the explanation). The main reason is that there was a gap between the Sporer and Maunder minimums which saw relatively high solar activity, and yet that time period (16th to early 17th century) was also definitely part of the Little Ice Age, with no apparent moderation of the cooling during that time.

In terms of the end of the LIA, this is where our early instrumental data starts to be useful:

In a way I suppose it depends on your definition - most of the models show an early warming trend around the late 19th century, which then reverses, then a more sustained warming as you say beginning in the early 20th century. The mid-20th century reversal is now well understood (due to aerosol emissions, and it also used to look bigger than it was due to measurements being more focused over the Northern Hemisphere at that time).

My suggestion would be that the LIA was clearly on its way out by the late 19th century, but then there were three VEI 6 eruptions (Krakatoa 1883, Santa Maria 1902, Novarupta 1912) all of which had shorter term impacts (which can be seen quite clearly on the chart above), so it's somewhat debatable whether you'd call it a separate period of cooling or still part of the LIA - more of an academic point really.

In terms of the trends in the Little Ice Age, the other reason to look at volcanoes more so than solar is the lack of a truly global extent of the changes, or at least that on a global scale they were relatively modest. The major effect appears to have been in the Northern extratropical region (30N to 90N), as the chart by Ljungqvist suggests in my original post, though there were likely smaller global effects (though not meeting the gold-standard 95% statistical significance) on the order of perhaps a quarter of a degree or so. But the 'classical' pattern is best seen in Ljungqvists graph, which clearly shows a distinct Roman Warm Period, then a cooling trend, then a Medieval Warm Period, then a Little Ice Age, and then the modern warming, as being the pattern for the Northern Hemisphere away from the tropics.

[Midlands Ice Age]

20 minutes ago, WYorksWeather said:

The postulated reason for cooling would indeed be land use changes - e.g. the human practice of things like clearing land for farming etc. could have already been having smaller impacts over that time period, and the large reduction in the population would have slowed down or even reversed that process (large areas presumably going back to nature). That of course is only a hypothesis, and is very much contested in the scientific community as an explanation. You're right that it is far too soon for any sort of emissions-based warming, and in fact even into the early 20th century this was quite modest relative to today's levels (at most 5-10% of the modern emissions) and would of course have been far lower still in the European late medieval and Renaissance/Enlightenment periods.

In terms of the solar / volcanic balance, there is a good reason for saying that volcanic is a major, or even dominant factor (though of course it's unlikely to be 100% of the explanation). The main reason is that there was a gap between the Sporer and Maunder minimums which saw relatively high solar activity, and yet that time period (16th to early 17th century) was also definitely part of the Little Ice Age, with no apparent moderation of the cooling during that time.

In terms of the end of the LIA, this is where our early instrumental data starts to be useful:

In a way I suppose it depends on your definition - most of the models show an early warming trend around the late 19th century, which then reverses, then a more sustained warming as you say beginning in the early 20th century. The mid-20th century reversal is now well understood (due to aerosol emissions, and it also used to look bigger than it was due to measurements being more focused over the Northern Hemisphere at that time).

My suggestion would be that the LIA was clearly on its way out by the late 19th century, but then there were three VEI 6 eruptions (Krakatoa 1883, Santa Maria 1902, Novarupta 1912) all of which had shorter term impacts (which can be seen quite clearly on the chart above), so it's somewhat debatable whether you'd call it a separate period of cooling or still part of the LIA - more of an academic point really.

In terms of the trends in the Little Ice Age, the other reason to look at volcanoes more so than solar is the lack of a truly global extent of the changes, or at least that on a global scale they were relatively modest. The major effect appears to have been in the Northern extratropical region (30N to 90N), as the chart by Ljungqvist suggests in my original post, though there were likely smaller global effects (though not meeting the gold-standard 95% statistical significance) on the order of perhaps a quarter of a degree or so. But the 'classical' pattern is best seen in Ljungqvists graph, which clearly shows a distinct Roman Warm Period, then a cooling trend, then a Medieval Warm Period, then a Little Ice Age, and then the modern warming, as being the pattern for the Northern Hemisphere away from the tropics.

I am currently closely watching the southern hemisphere too as part of any climatic change currently taking place in terms of the Hunga Tonga Southern Pacific Volcano. This now seems  to be having a more global impact in terms of the worlds sea ice area and extent levels, which, if correct, may also give us an insight into the volcano impacts on our climate and surroundings.

I am currently composing a post for the N H Arctic sea ice thread, explaining the latest details that seem to be becoming apparent, but if you also consider when the waters covered 90% or more of earth, (guessing at least 50% of the latest 3 billion of the earth's history) then the eruptions of many underwater volcanos  could well have had major impacts on our historical climate.

It would seem that a not inconsiderable amount of the current warming could have been caused by the sudden huge increase of H2O into our stratosphere. This is something that was/is still  being researched, but looking more and more likely to have had  at least short - term effects. It might well have been that several could have erupted about the same time (relatively) under the huge ocean extent and this might well have affected the climate in a quick and sudden way. Probably more so  than the very small amounts of CO2 locked away at the time for earth bound  volcanos to emit, (going back before roughly a billion of years ago).

MIA

[WYorksWeather]

10 hours ago, Midlands Ice Age said:

I am currently closely watching the southern hemisphere too as part of any climatic change currently taking place in terms of the Hunga Tonga Southern Pacific Volcano. This now seems  to be having a more global impact in terms of the worlds sea ice area and extent levels, which, if correct, may also give us an insight into the volcano impacts on our climate and surroundings.

I am currently composing a post for the N H Arctic sea ice thread, explaining the latest details that seem to be becoming apparent, but if you also consider when the waters covered 90% or more of earth, (guessing at least 50% of the latest 3 billion of the earth's history) then the eruptions of many underwater volcanos  could well have had major impacts on our historical climate.

It would seem that a not inconsiderable amount of the current warming could have been caused by the sudden huge increase of H2O into our stratosphere. This is something that was/is still  being researched, but looking more and more likely to have had  at least short - term effects. It might well have been that several could have erupted about the same time (relatively) under the huge ocean extent and this might well have affected the climate in a quick and sudden way. Probably more so  than the very small amounts of CO2 locked away at the time for earth bound  volcanos to emit, (going back before roughly a billion of years ago).

MIA

Essentially, the key question is whether this year's warmth needs an additional explanation. I'll use Berkeley Earth since I follow their figures quite a lot, and though different datasets disagree on amounts of warming since pre-industrial, the disagreement between last year and this year will be very small. Berkeley Earth recorded 1.24C for 2022 above pre-industrial, and the estimate for the end of this year is around 1.5-1.6C. So, we have around 0.3C of warming to explain.

Underlying warming trends would give you 0.02 or 0.03C, and transition from moderate La Nina in 2022 to neutral in 2023 would give another 0.1C. You then have reductions in aerosol emissions, the 'weirdness' of this El Nino (coming from a triple-dip La Nina, which hasn't happened in over a century, so may be somewhat unusual in having a stronger rebound), and as you say Hunga Tonga.

These hypotheses will be relatively easy to test I think in hindsight - if we see an additional amount of warming that is permanent, it's likely reduction in aerosols that is the main culprit, but if we return to the previous slow and steady trend, or possibly even see a short-term flatline, then it's more likely El Nino or Hunga Tonga or both.

Here are things I would say with confidence:

• 2024 will be warmer than 2023
• No new record will be set after that for a few years
• A fresh set of records will be set coinciding with the next strong El Nino, likely in the late 2020s or early 2030s
• 2020-2029 will be warmer than 2010-2019
• 2030-2039 will be warmer than 2020-2029
• 2030-2039 will exceed 1.5C above pre-industrial in most global temperature datasets

[raz.org.rain]

An interesting detail about the little ice age is that it wasn't unremittingly cold. If I remember right, some of our record hottest CET annual averages come from this period. The best example being 1666, which saw a very dry and hot summer. This created the perfect cinder box conditions for the infamous Great Fire Of London to spread. 

[Midlands Ice Age]

 

WYorks

I think that the new CR Euro climate dataset looks more realistic for this current year than using Berkeley earth.. It is now showing a fall in the anomalies over the last month to more like the levels we might expect. This coincides with the changes that I have observed in the ice extent numbers....

It is still running high, but the extreme anomaly picked up during the previous 6 months,   does  now appear to be dropping.

Time is needed  before this is resolved.

[WYorksWeather]

12 minutes ago, Midlands Ice Age said:

 

WYorks

I think that the new CR Euro climate dataset looks more realistic for this current year than using Berkeley earth.. It is now showing a fall in the anomalies over the last month to more like the levels we might expect. This coincides with the changes that I have observed in the ice extent numbers....

It is still running high, but the extreme anomaly picked up during the previous 6 months,   does  now appear to be dropping.

Time is needed  before this is resolved.

Last post on this for tonight, but just to say that the Climate Reanalyzer dataset doesn't conflict with Berkeley Earth, it's just more up to date. Berkeley Earth like the other datasets e.g. HadCRUT5, GISTEMP, UAH, etc., is updated monthly, so if there is a drop in the anomaly in December it will be reflected. Also important to note that Berkeley Earth's estimate comes with uncertainty included, hence my use of 1.5C-1.6C for the annual anomaly, since we don't have December data yet. A lower December anomaly means it will be closer to 1.5C than 1.6C.

Anomalies from the Climate Reanalyzer charts are plotted relative to 1979-2000 as well, for reference (not sure if you were aware of that, as it's somewhat well hidden in the small print well below the chart), so they can't be directly compared anyway without some adjustment factor being applied.

Due to small differences in temperature trends in the late 19th century, estimates of warming since the mid-20th century are far better constrained than those plotted since 1850. Berkeley Earth's anomaly reads higher than most other datasets relative to 1850-1900, but I think you will find that when all the 2023 data is in, though the datasets will disagree on the 1850-1900 warming by around 0.2C from highest to lowest (excluding UAH, which follows a lagged trend and is somewhat different as it measures the entire lower troposphere rather than the surface), the agreement between them on the difference between 2023 and 2022 will be very close indeed, likely 0.05C or less.

When we get all the data in January, I may put a thread together, as it might be interesting, and of course we're moving well away from the original intended topic of this thread if we continue this discussion much further here!