March 27, 2007

Wildfires: The Results of Cyclic Oceanic Variations or Global Warming?

Filed under: Climate History

One of the common images associated with global warming is that that of a blackened and burned forest. The voice-over with such images either directly or indirectly links those images of wildfire to a worsening climate that is undoubtedly caused by us (see here for example). And as we all know, the modern 24/7 media loves to splash photos and imagery of any massive wildfires. Take for example those that have recently happened in such diverse places as Malibu, California and eastern Australia. Other massive widespread wildfires have swept western North America many times in the recent past, such as during 1996, 2000, and 2002. The implication—again sometimes spoken and sometimes not—is that greenhouse climate change is somehow responsible for these horrible calamities.

Given that graphic and much hyped scenario, one might be surprised to learn that most large-scale fire events in the western United States over the last five hundred years has been fundamentally the result of natural ocean climate cycles, and not global warming.

That is the conclusion reached by researchers from Colorado, Arizona, Montana and Argentina in a 2007 study published in the Proceedings of the National Academy of Sciences. Lead author Thomas Kitzberger and many colleagues from institutions such as the University of Colorado, the University of Arizona, the Universidad Nacional del Comahue and other places have used extensive tree-ring analyses to determine the underlying causes of these widespread and episodic wildfires.

The researchers examined over 33,000 fire events from nearly five thousand fire-scarred trees, primarily ponderosa pine and Douglas fir, throughout the western portion of North America. The team used fire data that are archived in the International Multiproxy Paleofire Database and are maintained by the National Oceanic and Atmospheric Administration. The exact calendar year of the fire event was obtained by noting the specific year of the tree ring record in which a fire scar occurred. The researchers investigated 238 sample sites throughout the western half of North America with an average sample of twenty trees per site. After statistical analysis (for example, cluster analysis) of all the data to determine the main geographical patterns of fire across the region, the researchers reduced the dataset into representative components (specifically through principle component analyses) that characterized the variations in the whole network.

The scientists then linked these representative components with measures of drought (specifically the Palmer Drought Index) and corresponding records of sea-surface temperature. They concentrated on three specific oceanic climate phenomena: First, they obtained records of the sea-surface temperatures in the central equatorial Pacific in order to study the effects on wildfires by the well-known periodic Pacific climate phenomenon called ENSO or Southern Oscillation/El Nino. ENSO is a phenomenon that scientists have charted throughout the past and have determined occurs with a roughly 5-7 year period. Another strong, but longer-term, ocean climate phenomenon has also recently been identified in the Northern Pacific and is called the PDO or Pacific Decadal Oscillation. The PDO generally over the century has created variations in sea-surface temperatures in the Northern Pacific in multi-decadal (e.g., 20 year) periods. The third and final oceanic climate phenomena that the researchers studied is a strong ocean variation found in the North Atlantic Ocean termed the AMO or Atlantic Multidecadal Oscillation that operates on much longer periodicities of up to 60-80 years.

The researchers wrote, “Links between contingencies of ENSO and PDO, drought and wildfires have been noted in the western U.S. before, but only for subregions or using 20th century data. Furthermore, AMO-fire teleconnections and contingencies with ENSO and PDO previously have been shown for single study areas but have not been examined at regional scales.” Consequently, they took this massive database of 33,000 fire events derived from tree-ring records across the entire western North America and compared it to these ocean climate phenomena.

Not surprisingly, Kitzberger and colleagues found that, since the year 1550, fire events and droughts have been linked—the occurrence of drought and wildfire in the West have coincided. More importantly, they found these occurrences have been strongly tied to variations in the three ocean climate phenomena—ENSO, PDO and AMO. They found that the phases of the two Pacific climate phenomena, ENSO and PDO, were quite important in determined the frequency of fires in subregions across the West. In particular, they found that ENSO is quite important in determining regional wildfires on annual scales. But, given the recent massive wildfires across the West, they were most interested in whether widespread fires—extending across large areas of the West—were linked to these oceanic climate variations.

They found that the ocean temperature variations created by the Atlantic Multidecadal Oscillation are actually the dominant factors in whether wildfires were widespread across the West. They found that the warm phase of the AMO appears to be the critical factor in explaining the massive outbreaks of fire across the West—including the most recent ones. As the lead author noted, “The key issue is that the Atlantic Multi-Decadal Oscillation persists on time scales of 60 to 80 years, compared to just one year or a few years for El Niño.”

So does their study have any implications for the future? Can we tell something about upcoming widespread fire occurrence in the western portion of North America based on this extensive tree-ring analysis? The answer is yes—and the implications aren’t too good. Unfortunately, we are currently in the warm phase of the AMO (which has also been fingered as one the primary reasons for the enhance hurricane activity in the Atlantic Ocean since 1995) and, given the very long-term nature of this Atlantic ocean/climate phenomenon, we are likely to be in the AMO warm phase for quite some time to come. That suggests we will likely continue to see more and more media coverage of massive wildfires in the West … and the continued implication that somehow global warming is the cause of them—an implication that these respected tree-ring scientists are suggesting is not valid.


Kitzberger, T., Brown, P.M., Heyerdahl, E.K., Swetnam, T.W. and T.T. Veblen, 2007. Contingent Pacific-Atlantic Ocean Influence on multicentury wildfire synchrony over western North America. Proceedings of the National Academy of Sciences, 104 (2), 543-548.

No Comments

No comments yet.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

Powered by WordPress