By: Patrick W. Zimmerman

The destruction wrought by Hurricane Harvey in Texas is terrifying and, as has been helpfully pointed out by sources as antagonistic to each other as President Trump and the Mainstream Media, is unprecedented.

But (previously) unprecedented disasters are happening with increasing frequency.  And the trend of shorter and shorter time periods between “100 year” and “500 year” emergencies will almost certainly continue.

This.  This is the most obvious real-world consequence of anthropogenic climate-change.  This is how the warming Earth is a systemic danger to us.  Not to our grandchildren or society in some hypothetical future, but right here, right now.  Climate change is strongly correlated with weather volatility.  That is to say, it’s not just that as the world gets hotter, there will be less snow (though that’s true), but that the intensity and frequency of weather extremes will increase.  For example, climate change is correlated with:

• Stronger tropical cyclones¹ that particularly affect coastal wetlands (a critical component in natural and man-built flood and storm surge regulation).²
• Increased risk of flooding (both associated with tropical cyclones and unrelated, such as those caused by winter storms).³
• Longer and more frequent droughts (helping contribute to…)⁴
• More forest fires burning more acreage.⁵

Why does this matter so much?  Because disasters involve people (possibly including you) losing their lives and livelihood. For those of you who value money more than humanity, it also matters because they’re also expensive as hell, particularly when preparedness and disaster response depends on an aging infrastructure and disaster management plans made with out-of-date predictions.

Note: Article, database, and dashboard have been updated to include Hurricane Irma. Shockingly, the argument has not been invalidated.

The question

As the Earth warms, is climate volatility already showing up as an increase in the number and frequency of disasters in the US?

The short-short version

Yes.  Yes, it is.  We’re totally screwed.

FEMA has been more and more active, particularly over the last two decades.  The number of events within climate-correlated categories has risen precipitously, and it is clearly correlated with the warming of the planet as measured by average sea surface temperature (R² = 0.45 0.46, R = 0.67 0.68).

So get ready for the next Harvey or Irma or Katrina or Sandy.  And the one after that.

Hover for details. Select query to highlight.

The methodology

To avoid semantic arguments, we’re not going to debate what constitutes a disaster; FEMA has already conveniently defined that for us.  We’ll go with their criteria and include everything that has been formally declared a Major Disaster (DR coded event) or an Emergency (EM) since the government started the formalized nomenclature and its concurrent release of federal funds in 1953.⁶

First order of business was to adapt a shell script (thanks, Thor!) to query the FEMA server for every declaration made, extract just the major disasters and emergencies, then parse the resulting .csv by date, location, and cause.⁷

The historical data for average sea surface temperature (going back to 1880) is available from NASA.⁸

Disaster declarations since 1953

Hover for details. Select query to highlight.

It’s pretty terrifying to look at how much the rate of disaster declarations has spiked. Because of the way that FEMA declares disasters and emergencies, some of these (such as EM-3262 – New York Hurricane Katrina Evacuation) serve to measure an events severity rather than the quanity of events. A really major disaster will spawn multiple FEMA declarations, for aid, refugee resettlement, and the like, even outside the area directly affected by the storm. For those of you who are interested, yes the big spike in 2005 is Hurrican Katrina, and Hurricane Harvey has triggered 2 events: one disaster declaration (DR-4332 – Texas) and one Emergency (EM-3382 – Louisiana). Irma has sparked 9 (so far). Emergencies have been declared for Irma in the US Virgin Islands, Puerto Rico, Florida, South Carolina, Georgia, and with the Seminole Tribe (Florida). Irma has triggered Major Disasters in the US Virgin Islands, Puerto Rico, and Florida.

Climate-correlated v. non-climate related disasters

Hover for details. Select query to highlight.

In the green: climate change is positively correlated with an increase in frequency or severety. This category includes: blizzards and severe storms (and associated effects like avalanches, floods, landslides/mudslides, and tornadoes), el niño events, extreme temperature, fishery depletion, drought, and tropical cyclones (cyclones, hurricanes, typhoons, tropical storms, and tropical depressions), and wildfires.

The grey line represents all other disasters. This includes accidents (chemical waste/spills), infrastructure failure (dams, levees, and canals failing, power grid crises, etc), terrorist attacks, and seismic events (earthquakes, tsunamis, and volcanoes).

Short story: the climate-correlated disasters are both increasing in frequency and represent the vast majority of emergencies to which FEMA needs to respond.

Named storms

Hover for details. Select query to highlight.

How about just looking at tropical cyclones that make US landfall (at an intensity strong enough to elicit a federal disaster declaration)?

Yup, trending up, as well. It’s not just the blizzards, coastal storms, and surges that are going up, but storms big enough for us to need to name them are also hitting US territory with increasing frequency and intensity. Hurricanes, in particular, tend to create a cascade of FEMA declarations due to both their multi-state scope and occasional secondary declarations to help other states lend aid. The huge spike in 2005 includes no fewer than 47 separate emergency declarations (EM-3212-3158) to release federal money to other states to help deal with the evacuation, refugee crisis, and temporary housing caused by Katrina. One was issued for every state and DC except for Alaska and Hawai’i (for obvious logistical reasons), and Wyoming and Vermont (no idea).

The disaster map

Hover for details. Select query to highlight.

Every federal disaster declaration since 1953, in handy map form, sorted by state or territory (Guam, American Samoa, and the rest of the American Pacific islands are why typhoons and cyclones show up in the dataset.). The map is sortable by location, type, subtype, and name. Tribal councils are mapped onto the state in which their reservation primarily falls.

You see a pretty predictable mapping of type. Cyclones are concentrated in the South and along the Eastern Seaboard (including the Mid-Atlantic and New England probably more than you thought), as well as in the Pacific. Seismic events are pretty much confined to the Pacific Ring of Fire and its associated volcanoes and faults, plus Hawai’i. California actually does have an outsized number of wildfires that reach emergency status (there’s an entire separate FEMA category for firefighting aid for fires that don’t rise to the level of an emergency).

Surprising: Drought declarations (usually given to help agriculture through a sustained dry spell) are pretty evenly spaced throughout the US. With the exception of the Midwest and Texas. No idea what’s up with that.

The big table of all major disaster and emergency declarations

Hover for details. Select query to highlight.

For the detail-oriented among you, here’s all of them through Hurricane Harvey. Use the drop-down menus to search for names, locations, dates, and type.

What’s next?

In the context of this project, we’re exploring the idea of examining the monetary expense of all these disasters (measured in money disbursed by the Feds for each disaster declaration) as well as the human cost (in lives lost) of each one, as well, with the follow-on possibility of being able to create a reliable dollar-per-degreeSST or lives-per-degreeSST metric. It depends on the completeness and availability of government record-keeping, though, so still indeterminate at this time.The former (money) is likely to be easier than the latter (casualties).

In the context of real life? The smart money says, “another Atlantic hurricane making landfall before the end of the 2017 Atlantic Hurricane season. At least.” Yeah, it’s a downer, but this isn’t exactly something for which there’s a quick policy fix. The question isn’t how to stop global warming, but rather how to not let an already inevitable major problem spiral into a catastrophic one.

Dispensing with the political myth that climate science has “two sides” and that we need to “continue the review and analysis” of whether humans are causing global warming would be a start, though.

Notes:
  ¹ Knutson et.al, “Tropical cyclones and climate change”, Nature Geoscience, no 3 (2010): 157-63. doi:10.1038/ngeo779 ^
  ² Michener et.al, “Climate change, hurrricanes and tropical storms, and rising sea level in coastal wetlands,” Ecological Applications, no 7 (1997):770-801. doi:10.1890/1051-0761. ^
  ³ Hirabayashi et. al, “Global flood risk under climate change”, Nature Climate Change, no 3 (2013):816–21. doi:10.1038/nclimate1911. Also Milly et. al, “Increasing risk of great floods in a changing climate”, Nature, no 415 (31 January 2002):514-17. doi:10.1038/415514a.^
  ⁴ Carnecier et. al., “Widespread crown condition decline, food web disruption, and amplified tree mortality with increased climate change-type drought”, Proceedings of the National Academy of Sciences of the USA, vol 108 no 4 (2010): 1474–78. doi: 10.1073/pnas.1010070108. Also, including drought’s connection to political and social unrest: Kelley et. al., “Climate change in the Fertile Crescent and implications of the recent Syrian drought”, Proceedings of the National Academy of Sciences of the USA, vol 112 no 11 (2014):3241–46. doi: 10.1073/pnas.1421533112. ^
  ⁵ Kurz et. al, “Mountain pine beetle and forest carbon feedback to climate change”, Nature, no 452 (24 April 2008): 987-90. doi:10.1038/nature06777. Also: Flannigan et. al, “Impacts of climate change on fire activity and fire management in the circumboreal forest”, Global Change Biology, vol 15 no 3 (March 2009): 549-60. doi: 10.1111/j.1365-2486.2008.01660.x and Brown et. al, “The Impact of Twenty-First Century Climate Change on Wildland Fire Danger in the Western United States: An Applications Perspective”, Climatic Change, vol 62 no1–3 (January 2004): 365–88. doi: 10.1023/B:CLIM.0000013680.07783.de ^
  ⁶ DR-1 was the Warner-Robins Air Force Base tornado in Georgia, April 30, 1953, declared 2 days later.^
  ⁷ FEMA. Total number of declared disasters, by state and by year.^
  ⁸ NASA Goddard Institute for Space Studies. GISS surface temperature analysis.^