The Extreme Winter of 1783-84
The Extreme Winter of 1783-84
The Extreme Winter of 1783-84
Tuesday, March 15, 2011
***Update (3/30/2011)***: Our Laki paper has been chosen as an AGU Research Spotlight (pdf). The Editor’s Highlight is available here. The Earth Institute’s State of the Planet Blog has also posted an entry on our Laki work, as did several other sources.
And now a break from the synthetic world of pseudoproxies...
Our paper on the extreme winter of 1783-84 has just been published in Geophysical Research Letters. Rosanne D’Arrigo from Lamont’s Tree Ring Laboratory was first author on this study, which was actually motivated by the extreme winter events of 2009-10. For those of us living in the eastern United States, the 2009-10 winter was a season to remember. Record snowfall was recorded in many eastern states, while extreme cold was similarly widespread throughout the region. Similar anomalous winter conditions were also observed throughout much of Europe. Richard Seager, another coauthor on the current paper, studied the likely dynamical cause of these 2009-10 winter conditions and published his team’s findings last year. The gist of their results was that the winter conditions in the U.S. and Europe were caused by a combination of an El Niño event in the equatorial Pacific and a strong and persistent negative phase of the North Atlantic Oscillation (NAO). Such a combination can cause increased snowfall in the southeastern U.S. because of the southward-displaced storm track due to El Niño, and farther north in the U.S. (and in much of Europe) positive snow anomalies arise from the cold induced by a negative NAO. Such is the prevailing thinking about last year’s winter, but how unusual was this combination of El Niño and negative NAO?
Generally speaking, the combined occurrence of these two states in the ocean and atmosphere is not that uncommon. El Niño events occur every 2-8 years and the NAO oscillates between persistently negative and positive phases over the course of decades (the last period of persistent negative NAO events was in the 1960’s and 70’s). What may be most important for their influence on the extremeness of winters in the eastern U.S., however, is the combined strength of each of these states and it turns out that last year’s combination was the strongest in the modern record (extending back to the mid 1800’s). This is interesting in its own right, but we wanted to put the modern record in a longer historical context. What we found in our GRL paper was that the combined strength of El Niño and the negative NAO in 2009-10 was even more anomalous than previously thought: it trumped everything since the 1400’s. But that is not the end of the story.
To extend our understanding further into the past, we used multiple paleoclimatic records that provide estimates of El Niño occurrences and NAO states several hundreds of years into the past. These records were derived mostly from tree rings, but include some ice core and documentary evidence. The time series of these estimates are shown in the figure below and come from our paper.
The surprising result from the reconstructions above is not just the fact that 2009-10 was so anomalous, but that the second strongest combination of El Niño and a negative NAO was the winter of 1783-84, which turns out to be a very interesting period. This winter was roughly coincident with another important geological and climatic event: the eruption of the Icelandic volcano Laki, which comprised a series of punctuated eruptions beginning in the summer of 1783. Although the eruptions continued for eight months, the most explosive eruptions occurred at the beginning of the sequence in the 1783 summer. The consequences of the eruption were vast and severe, earning it the dubious distinction of being one of the worst natural disasters of the last millennium. Thousands died from gas fumes and famine, while a thick dust cloud spread over much of Europe causing additional extreme weather, crop failure and fatalities related to respiratory injuries. These catastrophes were concentrated over much of the summer and fall that year, but it turns out that the winter of 1783-84 was also very extreme in both Europe and the United States. It has long been thought that the Laki eruption was the cause of these winter conditions because of the radiative impact that large volcanic eruptions are known to have on on our climate. In fact, Benjamin Franklin is largely credited as first making the connection between volcanoes and climate in lectures following the Laki eruption. Nevertheless, while the impacts of Laki on the summer and fall conditions in 1783 are indisputable, our paper offers an entirely different explanation for the cause of the extreme winter that followed in the eastern U.S. and Europe. Instead of being caused by Laki, we argue that similar to 2009-10, the extreme winter of 1783-84 may in fact have been caused by a combined occurrence of an El Niño event and strong negative NAO.
Is this the last word? Perhaps not. There are unresolved questions about both the impacts of combined El Niño and negative NAO events, as well as the connection between these dynamic ocean and atmosphere modes and volcanoes. A big challenge with extreme events is, by definition, there just aren’t that many of them. It therefore is difficult to derive robust statistics for extreme events or to study the underlying dynamics that cause them. As a result, there may yet be some exciting new twists to this story, but for the time being we have proposed a new hypothesis that connects some recent extreme events with another very interesting period in history.
PAPER REFERENCE: D’Arrigo, R., R. Seager, J.E. Smerdon, A.N. LeGrande, and E.R. Cook (2011), The anomalous winter of 1783-1784: Was the Laki eruption or an analog of the 2009-2010 winter to blame?, Geophysical Research Letters, 38, L05706, doi:10.1029/2011GL046696. [Auxiliary Materials]
Laki Volcano, Iceland
Figure 1 (from D’Arrigo et al., 2011). (a) Normalized DJFM NAO reconstruction. (b) Normalized NAO Niño3 SST reconstruction. (c) Paleo‐Niño‐NAO index based on differencing normalized Niño‐3‐NAO SST. 2010 and 1784 values 1st and 2nd highest over the past six centuries (Table S1).