Could Climate Change Reduce the Frequency of Tracks Like Hurricane Sandy's?

We're used to seeing hurricane-battered beaches and flooded cities in Florida, North Carolina, and the Gulf Coast. But to see these images from the Jersey Shore and New York City in the wake of Hurricane Sandy was a shocking experience. New Jersey rarely gets hit by hurricanes, because it lies in a portion of the coast that doesn't stick out much, and is too far north. Hurricanes generally move from east to west in the tropics, where the prevailing trade winds blow that direction. But the prevailing wind direction reverses at mid-latitudes, due to the spin of the Earth, and the flow becomes predominately west-to-east. Hurricanes that penetrate to approximately Northern Florida's latitude typically get caught up in these westerly winds and are whisked northeastwards, out to sea. However, the jet stream, that powerful band of upper-atmosphere west-to-east flowing air, has many dips and bulges. These troughs of low pressure and ridges of high pressure allow winds at mid-latitudes to flow more in a south-to-north direction. Every so often, a trough in the jet stream bends back on itself when encountering a ridge of high pressure stuck in place ahead of it--a so-called "blocking high". According to NOAA's Climate Prediction Center, a "blocking high" near the longitude of Greenland (50°W) only occurs about 2% of the time in the fall, and can cause winds that flow from southeast to northwest over the Northeast U.S. It is this sort of unusual flow that sucked Sandy into New Jersey and allowed the hurricane to take the most perpendicular track into this section of coast of any tropical cyclone in the historical record (Hall and Sobel, 2013.) Using historical climate data, these scientists estimated that the return period of a Category 1 or stronger storm hitting New Jersey at such an odd angle was 1-in-700-years.


Figure 1. Tracks of all tropical storms and hurricanes to hit Southern New Jersey, 1851 - 2012. Hurricane Sandy had a track unprecedented in the historical record. Image created by TWC's Stu Ostro using data from NOAA/CSC.


Figure 2. Hurricane Sandy at 10:10 am EDT October 28, 2012. Image credit: NASA/GSFC.

Was climate change responsible for Sandy's unusual track?
Either Sandy was an extremely rare event, or else climate change has shifted the odds of such a track to make it more likely. A paper published on Monday in the Proceedings of the National Academy of Sciences by Elizabeth Barnes of Colorado State and co-authors, "Model projections of atmospheric steering of Sandy-like superstorms", argues that our best climate models project we should see a decrease in the type of steering patterns that brought Sandy to the coast at such an unusual angle. Of the 22 models used to formulate the 2013 IPCC report, 17 predict a decrease in the type of "blocking highs" responsible for Sandy's unusual track. Nineteen out of 22 of these models also predict that the jet stream will shift farther to the north, particularly in the fall. The authors argue that this jet stream shift will bring about a decrease in easterly winds south of Greenland, resulting in fewer Sandy-like storms hitting the Northeast U.S. However, Dr. Jennifer Francis, who has authored several studies linking Arctic sea ice loss to unusual jet stream patterns, noted in an email to me that "One of the strongest pieces of evidence for the study’s main conclusion is that easterly winds are projected to decrease in a large zone north of Newfoundland. The location of the strongest decreases, however, is north of the location of the block during Sandy, exactly in the region where stronger west winds would occur when blocks like this existed. This suggests that the pattern may actually cause an increase in unusually high pressures in the same location of the Sandy blocking high."


Figure 3. Jet stream winds at a pressure of 300 mb on October 29, 2012, as Hurricane Sandy approached the coast of New Jersey. Note that the wind direction over New Jersey (black arrows) was from the southeast, due to a negatively tilted trough of low pressure over the Eastern U.S. caused by a strong blocking ridge of high pressure over Greenland. Image credit: NOAA/ESRL.

Commentary
While the 2013 IPCC models used in the study are the best that we have, the uncertainties are very high in the sort of projected atmospheric changes the authors are analyzing. In the current climate, the models underestimate the frequency of the type of blocking high pressure systems that led to Sandy's unusual track. Thus, we should look with suspicion upon their predictions for a decrease in blocking highs in the future--something that Barnes et al. acknowledge in their paper. In addition, Arctic sea ice loss is occurring much faster than these models predicted; in 2012, September sea ice loss was more than 1 standard deviation below where these models predicted it should be. Thus, these models may be underestimating the influence of sea ice loss on hurricane steering flow in the North Atlantic. As I discussed in an April post, "Arctic sea ice loss tied to unusual jet stream patterns", three studies published in the past year have found that the jet stream has been getting stuck in unusually strong blocking patterns in recent years. These studies found that the recent record decline in Arctic sea ice could be responsible, since this heats up the pole, altering the Equator-to-pole temperature difference, forcing the jet stream to slow down, meander, and get stuck in large loops. The 2012 Arctic sea ice melt season was extreme, with sea ice extent hitting a record low. The 1-in-700 year Hurricane Sandy track could have had its odds boosted by the 2012 record sea ice loss, one can argue, based on this research. This research, however, is disputed by Dr. Barnes in a separate study just published in Geophysical Research Letters, "Revisiting the evidence linking Arctic Amplification to extreme weather in midlatitudes."

The Atlantic hurricane season has been getting longer in recent decades, in association with increasing ocean temperatures. A longer season gives the opportunity for more strong hurricanes to penetrate to the Northeast U.S. in late fall. Warmer ocean waters may also lead to an increase in strong hurricanes farther to the north, since cool ocean temperatures are a key reason why we see so few strong hurricanes affecting the Northeast. These influences would potentially offset any decrease in Sandy-like storms caused by fewer blocking highs forming in a future climate. Much more research is needed before we can be confident how climate change may or may not affect the tracks and frequency of future storms like Hurricane Sandy. One thing that is almost a sure thing: as global warming continues to cause sea levels to rise, the impacts of these storms will be worse as storm surge flooding penetrates farther inland.


Figure 4. Extent of Arctic sea ice predicted by the mean of 20 climate models used to formulate the 2013 IPCC report (thick red line); the pink area denotes plus or minus one standard deviation from the mean. The actual levels of sea ice (thick black line) fell below one standard deviation from what the model were predicting in 2012. The older predictions from the set of models used to formulate the 2007 IPCC report are shown in blue. Image credit: Stroeve et al. 2012, "Trends in Arctic sea ice extent from CMIP5, CMIP3 and observations", Geophysical Research Letters, DOI: 10.1029/2012GL052676.

Europe expected to see a large increase in Hurricane Sandy-like hybrid storms?
While the new study by Barnes et al. gives some hope that global warming might lead to fewer Sandy-like storms hitting the Northeast U.S., dangerous part-hurricane, part extratropical hybrid storms like Hurricane Sandy are expected to be an increasing threat for Western Europe by the end of the century due to global warming, said a team of scientists led by Reindert J. Haarsma of the Royal Netherlands Meteorological Institute. In a paper called "More hurricanes to hit Western Europe due to global warming", published in April 2013 in Geophysical Research Letters, the researchers describe the results from runs of a high-resolution (25 km grid spacing) climate model based on the European Centre for Medium Range Weather Forecasts (ECMWF) numerical weather prediction model. The model predicts that the breeding ground for Atlantic hurricanes will shift approximately 700 miles eastwards as the oceans warm this century. Hurricanes which form farther to the east can spend more time over warm tropical waters before turning north and northeast towards Europe, increasing the odds that these storms will have hurricane-force winds upon arrival in Europe. The model showed that wind shear will change little in the region over the coming decades, resulting in a large increase in storms with hurricane-force winds affecting Western Europe. Most of the these storms will not be hurricanes upon arrival in Europe, but will be former hurricanes that have transitioned to extratropical storms with hurricane-force winds. As we saw with Hurricane Sandy, these hybrid storms can be extremely dangerous. Summed over Norway, the North Sea, and the Gulf of Biscay, the model found that the number of hurricane-force storms in August - October increased by over a factor six, increasing from an average of two such storms in the current climate to thirteen per year by 2100. Almost all of these future Western European hurricane-force storms were predicted to originate as hurricanes or tropical storms in the tropics. The researchers conclude that "tropical cyclones will increase the probability of present-day extreme events over the North Sea and the Gulf of Biscay with a factor of 5 and 25 respectively, with far reaching consequences especially for coastal safety."

References
Barnes, E.A, L.M. Polvani, and A.H. Sobel, 2013, "Model projections of atmospheric steering of Sandy-like superstorms", PNAS September 3, 2013, doi: 10.1073/pnas.1308732110

Haarsma et al., 2013, More hurricanes to hit Western Europe due to global warming, Geophysical Research Letters, DOI: 10.1002/grl.50360

Hall, T.M., and A.H. Sobel, 2013, "The impact angle of Hurricane Sandy’s New Jersey landfall," Geophysical Research Letters 40:23122315.

"Europe expected to see a large increase in Hurricane Sandy-like hybrid storms", my April 2013 blog post.

"Why did Hurricane Sandy take such an unusual track into New Jersey?" my October 2012 blog post.

Arctic Warming May Not Be Altering Jet Stream: Study by Andrew Freedman of climatecentral.org, analyzing Dr. Barnes' paper disputing the link between Arctic sea ice loss and changes in the jet stream.

Quick update on 97L
The tropical wave that moved through the Lesser Antilles Islands on Tuesday morning (Invest 97L) is showing increasing signs of organization. It seemed like an atmospheric switch got thrown early this afternoon, and the storm began spinning up. Satellite loops show that 97L has developed some respectable low-level spiral bands, and we can see upper-level outflow channels opening to both the south and the north. There is no sign of a well-organized surface circulation, though, and dry air is still hampering the storm, as evidenced by surface-based outflow boundaries coming out of some of 97L's thunderstorms on its northwest flank. Upper level winds are favorable for development, with wind shear a low 5 - 10 knots, and an upper-level anticyclone overhead. In their 2pm EDT Tuesday Tropical Weather Outlook, NHC bumped up the 2-day odds of development to 30%. I put these odds at 50%, and Puerto Rico and the Eastern Dominican Republic can expect 3 - 6 inches of rain from 97L on Wednesday and Thursday. I'll be back Wednesday morning with an update.

Jeff Masters