Hurricanes & tornadoes in a warmer world

FSU Coastal & Marine Lab

Transcript

1. Hurricanes & Tornadoes in a Warmer World James B. Elsner (@JBElsner) Chair, Geography Department, Florida State University July 13, 2017 FSU Coastal & Marine Laboratory
2. How Are Hurricanes Changing?
3. Problem We know what hurricanes are like today. When, where, how often. But what about the future? Greater risk? No simple ways to get answers. Theory is limited. We don’t have a theory of climate. And we don’t know everything about hurricanes. Models don’t adequately represent the atmosphere or the ocean. And they don’t adequately resolve hurricanes. Data don’t go back far enough. They vary in quality. Social media won’t help unless you’re interested in mere opinions and bickering. Solution: Use available theories (thermodynamics, statistics) together with quality data.
4. Thermodynamics 1 1Photo Credit: Popular Mechanics
5. Thermodynamics Maximum potential intensity (MPI) MPI ∼ SST To BLf (SST) MPI is the highest wind speed (rotational) in units of meters per second. SST is the ocean temperature at the surface, To is the temperature at the top of the hurricane and BLf (SST) are heat fluxes involving potential temperature of saturated air. The heat fluxes depend on SST. Developed by Kerry Emanuel at MIT.
6. Statistics Extreme value theory (EVT) is a statistical theory that estimates the risk of extreme, rare events. Suppose we record the highest wind speed (m s−1) from 10 consecutive hurricanes. 34.5, 44.2, 57.5, 33.8, 67.8, 38.2, 41.5, 71.2, 61.0, 49.1 We order the values from lowest to highest. 33.8, 34.5, 38.2, 41.5, 44.2, 49.1, 57.5, 61.0, 67.8, 71.2 This tells us that 20% of the hurricanes have winds that exceed 61 m s−1 and 10% have winds that exceed 67.8 m s−1. EVT uses these quantile wind speeds to work out a theoretical highest possible wind speed, which we will call the limiting intensity (LI).
7. Statistics 0 2 4 6 8 10 20 40 60 80 Wind speed (rotational) [m/s] Frequency a 30 40 50 60 70 80 q q q q q q q q q q q q q q q q q q q q q 2 5 10 20 50 100 Return period [yr] Wind speed [m/s] b Limiting intensity is indicated by the red line in panel b.
8. Limiting intensity (LI) is a statistical quantity that we can use to compare with Kerry Emanuel’s theoretical maximum potential intensity (MPI). How should we make this comparison? Absolute values of LI are not as important as how LI changes with ocean temperature (SST). How can we estimate how LI changes with SST?
9. Data Hurricanes move over oceans that have varying temperatures. We need to match ocean temperature with hurricane intensity. Sea Surface Temperature (oC) 100°W 80°W 60°W 40°W 20°W 10°N 20°N 30°N 40°N 50°N 100°W 80°W 60°W 40°W 20°W 10°N 20°N 30°N 40°N 50°N 16 18 20 22 24 26 28 30
10. Spatial Grids We collect information about the hurricane in grids. The grids cover areas where hurricanes move. Hexagons more efficiently tile the region. a Hurricane counts 1 2 b Hurricane counts 1 40 80 120 160 200
11. a Number of hurricanes (1981−2010) c d 15 20 25 30 35 b Observed highest intensity [m s−1] c d 40 50 60 70 80 c Wind speed [m s−1] Percent of total 0 10 20 30 20 30 40 50 60 70 80 d Wind speed [m s−1] Percent of total 0 10 20 30 20 30 40 50 60 70 80 e Return period [yr] Return level [m s−1] 30 40 50 60 70 80 1 10 100 1000 • • • • • • • • • • • • • • • • • • • • uc LIc f Return period [yr] Return level [m s−1] 30 40 50 60 70 80 1 10 100 1000 • • • • • • • • • • • • • • • • • • •• • • • • • • • ud LId
12. Sea Surface Temperature (SST) & Limiting Intensity (LI)
13. 50 60 70 80 90 q q q q q q q q q q q q q q q q q q q q 26.5 27.0 27.5 28.0 28.5 29.0 Sea surface temperature [°C] Limiting intensity [m s−1] We estimate the sensitivity to be 8 ± 1.2 m s−1 K−1 (s.e.) for hurricanes over seas hotter than 25◦C.
14. Why Is This Important? The strongest hurricanes are getting stronger at a rate of 1 m s−1/decade.
15. 1e+07 1e+09 1e+11 q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q 20 40 60 80 Wind speed [m/s] Economic Damage [billion USD (2012)] U.S. 1e+07 1e+09 1e+11 q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q 20 40 60 80 Wind speed [m/s] Economic Damage [billion USD (2012)] Gulf coast 1e+07 1e+09 1e+11 q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q 20 40 60 80 Wind speed [m/s] Economic Damage [billion USD (2012)] Florida 1e+07 1e+09 1e+11 q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q 20 40 60 80 Wind speed [m/s] Economic Damage [billion USD (2012)] East coast Losses increase by 5% per 1 m s−1 increase in wind speed.
16. Summary We can understand what hurricanes might be like in the future by combining theory with data. It is likely that the strongest hurricanes will get stronger at about 8 m s−1 per degree of ocean warming. This amounts to about 1 m s−1 per decade at the current rate, which translates to about a 5% increase in loss per decade independent of exposure. What other factors affect this sensitivity? Shear, upper-level temperature, etc.
17. How Are Tornadoes Changing?
18. Problem: Annual Counts Provide Only a Limited View A strict focus on annual counts limits how we might understand a changing tornado climate. The fact that annual (E)F1+ tornado counts are not trending tends to stifle discussion about climate change & tornadoes. The absence of trends in tornado occurrence does not imply that tornado climate is stationary. The number of days with tornadoes or where they occur provide additional analytics for understanding how tornadoes might be changing.
19. U.S. Tornado Dataset No theory yet. The U.S. Storm Prediction Center (SPC) maintains the best available record of tornadoes in the United States compiled from National Weather Service (NWS) Storm Data publications and reviewed by the NCDC. Improved observational practices lead to an increase in the number of reported weaker tornadoes. The damage scale used to rate tornadoes was first published in 1971 and adopted for rating tornadoes in the near aftermath in 1973. Brief tornadoes may go undocumented in places with few people or limited communication infrastructure. Population bias.
20. Population Bias 0.8 1.0 1.2 1.4 0 20 40 60 Distance to Nearest City Center (km) Tornado Reports/10 km2
21. Decreasing Bias 1961−1970 1962−1971 1963−1972 1964−1973 1965−1974 1966−1975 1967−1976 1968−1977 1969−1978 1970−1979 1971−1980 1972−1981 1973−1982 1974−1983 1975−1984 1976−1985 1977−1986 1978−1987 1979−1988 1980−1989 1981−1990 1982−1991 1983−1992 1984−1993 1985−1994 1986−1995 1987−1996 1988−1997 1989−1998 1990−1999 1991−2000 1992−2001 1993−2002 1994−2003 1995−2004 1996−2005 1997−2006 1998−2007 1999−2008 2000−2009 2001−2010 2002−2011 0.1 0.2 0.3 0.1 0.2 0.3 0.1 0.2 0.3 0.1 0.2 0.3 0.1 0.2 0.3 0.1 0.2 0.3 0 20 40 60 0 20 40 60 0 20 40 60 0 20 40 60 0 20 40 60 0 20 40 60 0 20 40 60 Distance From Nearest City Center (km) Tornado Reports/100 km2 per Decade
22. Counts & Days 300 400 500 600 700 800 900 1960 1980 2000 Year Number of Tornadoes a 75 100 125 150 175 1960 1980 2000 Year Number of Tornado Days b The fact that annual (E)F1+ tornado counts are not trending tends to stifle discussion about climate change & tornadoes, but the absence of trends in frequency does not necessarily imply that tornado climate is stationary.
23. Tornado Days N = 4 N = 8 N = 16 N = 32 20 30 40 50 60 70 10 20 30 4 8 12 0 2 4 6 1960 1980 2000 1960 1980 2000 Year Annual Number of Days With At Least N Tornadoes
24. Probability of Big Days N = 4 N = 8 N = 16 N = 32 0.3 0.4 0.05 0.10 0.15 0.20 0.25 0.05 0.10 0.00 0.02 0.04 1960 1980 2000 1960 1980 2000 Year Probability of a Day With At Least N Tornadoes On Days With At Least One Tornado
25. Percent Change in Probability of a Big Day 1954−2013 1974−2013 1984−2013 1994−2013 0 5 10 15 0 5 10 15 4 8 16 32 4 8 16 32 N Percent Change in Annual Probability of a Day With At Least N Tornadoes On Days With At Least One Tornado a The change is computing using a logistic regression of annual probability onto year. The change is percent per annum.
26. Proportion Occurring on Big Days N = 4 N = 8 N = 16 N = 32 0.4 0.5 0.6 0.7 0.1 0.2 0.3 0.4 0.5 0.0 0.1 0.2 0.3 0.00 0.03 0.06 0.09 1960 1980 2000 1960 1980 2000 Year Proportion of All Tornadoes Occurring On Days With At Least N Tornadoes
27. Large-Scale Hypothesis An explanation for the increasing efficiency (more big tornado days) is simply larger areas favorable for tornadoes. We call this the large-scale hypothesis because it hints at the involvement of larger scale dynamical factors like wind shear as the cause. To examine evidence for this hypothesis, here we consider changes to the spatial dimensions of tornado occurrences. But what do we mean by a tornado cluster?
28. Example: May 8, 2008 Cluster Size (Number of Tornadoes) 7 11 a
29. All Clusters (1954–2013) Cluster Size (Number of Tornadoes) 50 100 b
30. No Trends 1 2 3 4 5 1960 1980 2000 Year Mean Number of Clusters Per Tornado Day a 100 200 300 400 1960 1980 2000 Year Total Area of Clusters (10000 sq km) b
31. Significant Increase in Tornado Density N = 4 N = 8 N = 16 N = 32 0 1 2 3 4 0 1 2 3 4 1960 1980 2000 1960 1980 2000 Year Annual Tornado Density Per Cluster (Tornadoes/10000 sq km)
32. Summary It appears that the risk of big tornado days with densely packed clusters of tornadoes is increasing. The increasing density of tornado occurrences within clusters suggests that the explanation for an increasing proportion of tornadoes occurring on big days might involve local-scale thermodynamics. We hypothesize that increases in both CAPE, driven by increases in low-level moisture, and CIN, driven by warming aloft, could lead to fewer days with tornadoes and to smaller, but more active, areas of severe convection on days with tornadoes.
33. Are Tornadoes Getting Stronger? Problem An estimate of how strong a tornado can get requires a continuous scale of intensity. The EF damage scale is categorical. We can count the number of tornadoes by EF category. But a time-series plot of the number of tornadoes does not answer the question: Are tornadoes getting stronger?
34. Tornado Path
35. Path Area by EF Rating Table: Damage path statistics.2 Rating N Area (km2) Mean Median EF1 7735 1.2 0.3 (0.1,1.1) EF2 2224 5.3 1.8 (0.5,4.9) EF3 650 18.5 9.0 (2.9,20.7) EF4 145 45.5 21.2 (7.4,50.5) EF5 14 103.1 71.6 (41.8,127.0) 2Data are based on all reported tornadoes in the United States (1994–2016). N is the sample size. Lower and upper quartile values are given in parentheses.
36. More Tornado Energy 3 4 1 2 1995 1999 2003 2007 2011 2015 1995 1999 2003 2007 2011 2015 0.0 0.2 0.4 0.6 0 5 10 0.00 0.03 0.06 0.09 0.12 0 1 2 3 4 Year Approximate Energy Dissipated [PW]
37. Longer Paths (Florida) 1995 2000 2005 2010 2015 .1 1 10 100 Path Length (km)
38. Final Thoughts Single metrics of storm activity can be misleading in climate-change studies. How often and how strong are the two canonical components of storminess. Analyzing frequency independently from intensity illuminates a broad theoretical space of storm climate variability. With typhoons in the Pacific and tornadoes in the U.S., the finger print of climate change appears to be fewer, but stronger.
39. Year Relative Magnitude above Below normal Activity Global warming influence