All gone by the year 2020: is observed glacier melt the basis for the prediction?

This is part 6 in the series on the prediction that glaciers in Glacier National Park will be gone by 2020. You might want to see to part 1, part2, part 3, part 4 and part 5 if you haven’t already.

In a previous post, I walked down memory lane to the time I was about 19 years old. In this post, I will go back to the more tender age of 14 and my first trip abroad. It was a trip to Switzerland on the health insurance plan of my parents. Its intention was to give the working class youth the opportunity to breathe in the pure mountain air of Switzerland. We were a bunch of kids of the same age and, as you would expect, there were a lot of outdoor activities like sports that we could chose from.

Not being that much into sports and a more nature-minded boy, I often joined the hiking group that did nature walks in the vicinity of the center were we stayed over. At one point, there was the opportunity to see a glacier. I had learned about glaciers in school and was eager to see one. It was not that close by and after a stiff walk we arrived at the glacier. Boy, was I disappointed. We saw some melting ice chunks and a small patch of ice stretching along the mountain. Although it was nice to see some ice in that time of the year, it was a real anti-climax.

One of the guides then said that a lot of glaciers were shrinking and that this glacier was no exception. This was in the mid 1970s.

I was reminded of this scene from my youth when reading this paragraph in the Hall & Fagre paper “Modeled Climate-Induced Glacier Change in Glacier National Park, 1850–2100” (my emphasis):

On the basis of tree-ring analysis in the forest fronting the Agassiz and Jackson Glaciers, Carrara and McGimsey (1981) estimated that within Glacier National Park the maximum glacial advances during the Little Ice Age occurred just before 1860. Retreat rates derived from their tree-ring data showed that before 1910 glaciers retreated at a modest rate (< 7 m per year). That rate increased dramatically between 1917 and 1926, reaching more than 40 m per year. Above the tree line, Carrara and McGimsey used terminal moraines, naturalists’ notes, photographs, and park records to deduce that the glaciers retreated rapidly (> 100 m per year) between 1926 and 1932 and continued to retreat at more than 90 m per year until 1942. This period of accelerated retreat corresponds to a period of above-average summer temperatures in the climatic record of the region (figure 4). After the mid-1940s the rate slowed, but ablation continued.

That was an interesting read. Also the glaciers in Glacier National Park were receding well before the 1970s and apparently even had their biggest melt rate in the mid 1920s and 1930s. It was also interesting because actual numbers were used. There was less than 7 m/year from just before 1860 until 1910, more than 40 m/year between 1917 and 1926, then more than 100 m/years until 1932, 90 m/year until 1942 and slowing afterwards. This period of accelerated melt occurred well before we put significant amounts of CO₂ in the atmosphere.

However, the paper stayed rather vague about what happened after 1942. It only mentioned that there was a slowdown after 1942, but not what the actual rates were between 1942 and the year that the paper was published (2003). Wondering about the current rate, I searched for the missing numbers, but in vain.

Luckily, I found historical data on the area of the named glaciers in Glacier National Park. The data is not optimal though. Only the area of the glaciers was measured and there were only five measurements (“just before 1860”, 1966, 1998, 2005 and 2015). The periods in between the measurements were also vastly different (116, 32, 7 and 10 years), making a comparison rather difficult. There are however some interesting things to extract from it.

A first thing is that looking at the number of glaciers, I found an explanation for the difference between the number of active glaciers reported around 2010 (25) and now (26). I was wondering about this in the first post in this series. One of my guesses was that one glacier might have been below the threshold of 25 acres and by growing came above that threshold in a more recent measurement. However, I could not find a glacier that was been measured below the threshold in 2005 and that was measured as more than 25 acres in 2015. The explanation for the difference of active glaciers seems to be that two glaciers retreated in two separate masses. I already knew about the Blackfoot and the Jackson glaciers, but also the Grinnell glacier separated in two masses (the Grinnell and the Salamander glaciers). If those were counted as two separate masses, then one arrives at 26. Otherwise it would be 25.

A second thing that became clear from the area information is that the glaciers in Glacier National Park had experienced a significant area loss since 1850. At closer sight, those glaciers lost more than half of their area by 1966. According to the 2003 paper, most of that loss was in the 1920s-1930s-beginning 1940s. Apparently, it seems that those glaciers were already well on their way on death row when we started to put significant amounts of CO₂ in the atmosphere (1950s).

Last, but not least, something that I wanted to look more closely at is the retreat in a more recent time, more specifically in relation to the 2020 prediction that was made in 2009. There was some controversy about the reason why the 2020 revision seemed necessary. Most articles describing the decision claimed that observational evidence of melt was the direct cause. For example this was given as the reason for the 2020 estimate as mentioned in an interview of a colleague of Fagre:

After publication of that report, field observations showed glacier melt to be years ahead of the projections, causing scientists in 2010 to revise their “end date” to 2020.

It somehow suggests that it was the observation of glacier melt between the report was published (2003) and 2010 that led to the 2020 estimate (it was in fact March 2009 when the 2020 estimate was broadcast). However, Fagre told the journalists something else: that the global temperature increase projections from the IPCC were used as input in their model, but the actual observed temperature increase in the park to be twice as great and therefor a revision to 2020 (from 2030) was needed. Now we know that the 2020 estimate is found wrong, this theory of increasing melt in that period seems very unlikely.

So I wondered whether any increasing area loss would show in the historical data just before the end date revision was made? Meaning, is there an increasing loss when comparing 1966 → 1998 with 1998 → 2005?

When I look at the average annual decrease rate between 1966 and 1998, then I find that there was an average loss of 0.88% per year. Between 1998 and 2005, that number was 0.76% per year. Contrary to what I expected, the area loss did the opposite and decreased slightly in the period just before the revision was made. That does not bode well for the theory of “field observations of glacier melt years ahead of the projections”. That would mean that there would have been a strong decrease in the melt rate between 2005 and 2020. This seems not to be the case. The next measurement in 2015 showed that the average melt loss reverted back to the rate between 1966 and 1998.

But then, maybe we should specifically look at the two glaciers that were investigated in the 2003 paper that put forward the 2030 estimate? These two glaciers are the Blackfoot and Jackson glaciers. When I do that, then the difference is even more profound. The Blackfoot glacier even grew slightly (from 1,625,124 m² measured in 1998 to 1,630,173 m² in 2005. The rate went from a loss of 0.37% per year between 1966 and 1998 to a very tiny increase of 0.04% per year between 1998 and 2005. The Jackson glacier went from a decrease rate of 1.41% per year to a decrease rate of 0.15% per year in the same period (linear trend):

That is a decrease in the rate since the measurement of 1966 (Blackfoot glacier) and 1998 (Jackson glacier). What was not know in 2005, is that this slower trend continued, shown by the measurement of 2015. If that trend would persist until present, then that glacier would not be gone any time soon (linear trend):

But then, if the two studied glaciers would unlikely be gone by 2020, how many of the other glaciers are projected to disappear based on the (then) known data until 2005? To figure that out, I entered the other values in Calc, calculated the (linear) trend line for the data between 1850 and 2005 and made the projection for 2020 (linear trend):

I ended up with two glaciers that would projected to be gone by 2020: the Boulder glacier and the Thunderbird Glacier. However, the Boulder glacier already dropped below the 25 acres threshold) somewhere between 1966 and 1998, therefor not considered an active glacier since many decades by then.

The Thunderbird glacier was, just barely, considered an active glacier in 2005. but it also has a slowing trend in recent times. It started as a rather large glacier according to the measurements of 1850 and lost 7/8^th of its area before the measurement of 1966 and, based on the later measurements, the decrease went much more slowly, it almost flatlined between 1966 and 2005 (linear trend):

Considering that flat trend, it should be clear that in 2005 it should be less likely that this glacier would be gone by 2020 . Later, based on 2015 measurements, that trend continued at the same rate, so the Thunderbird glacier would unlikely be gone by 2020, unless there was a drastic change from 2015.

That is still zero glaciers gone by 2020 according to the data that was known at the time of the prediction in 2009.

But then, their definition of “glaciers gone” could be all “glaciers smaller than 25 acres” (active glaciers). There are 11 glaciers that would be projected smaller than 25 acres in 2020, but only two were still considered active in 2005: the Two Ocean glacier and the Whitecrow glacier. The former had the same slowing down of the melting trend (since 1998). If that decrease rate stayed the same, then the Two Oceans glacier would also not be projected gone by 2020. The Whitecrow glacier however was not much above the threshold in 2005 and it had a much less pronounced slowing at the end, so that glacier would be projected below the threshold of an active glacier in 2020, based on its trend until 2005 (also based on the 2015 data, it scraped just above the threshold, so is very likely to be below in later measurements).

I think it is safe to conclude that observed melting between 2003 and 2009 is not the cause of the prediction that “all glaciers would be gone by 2020” (even if we would define “gone” as “less than 25 acres”). At least when it comes to the area measurements and based on what was known by 2005. Something else seemed to be the trigger. Then the explanation of Fagre (that it was the, ahem, correction from global to local temperature increase) that was the more likely reason to make the revision.

Closing, although the data is far from optimal and not suitable for the purposes I originally wanted to use it for, I learned a lot by looking at it.

Go to Part 1 | Part 2 | part 3 | Part 4 | part 5 | part 6 | part 7

Advertisement