Predictions and Preparedness – How Much Stock Should We Put in Preseason Hurricane Forecasts?

Predictions and Preparedness – How Much Stock Should We Put in Preseason Hurricane Forecasts?

June 5, 2018

June 5, 2018 - June 1 marked the official start to the Atlantic hurricane season, which runs until November 30 in the Atlantic Ocean, Caribbean Sea, and the Gulf of Mexico and peaks in mid-August to late October. Despite the apparent early start to the season, with subtropical storm Alberto forming in the Gulf of Mexico before being downgraded to a subtropical depression over the Southeast US, most initial forecasts predict an average or slightly above average season for 2018.

To the amateur climatologist, an “average or slightly above average season” may sound unclear, or even indecisive on the part of forecasters. Uncertainty is not indicative of bad science or of ambivalent outlook authors, though – rather, it’s a manifestation of incredibly complicated modeling that underlies early hurricane season forecasting. And it only underlines the need to be prepared for the worst early on, regardless of predictions.   

The 2018 preseason forecast roundup

Hurricane season averages are based on the 30-year average from 1981 to 2010. Under this definition, the average season includes 12 named storms and six hurricanes, three of which qualify as category 3 or higher and so are considered “major.”

Early forecasting is susceptible to several atmospheric uncertainties that can upset even the best models’ predictions. These uncertainties include calculations for El Niño and La Niña, water surface temperatures, and other weather patterns that are all inherently difficult to predict. In other words, early season model forecasts are predictions based on other predictions, which generates a lot of room for error. NOAA and other forecasters are well aware of this, and acknowledge the challenges inherent to their forecasting efforts. To me, early season outlooks are more valuable as a scientific exercise than a tool to inform planning for the season.

The following table summarizes three preseason 2018 predictions from a few of the most-cited forecasters in the game.

Forecaster

Named storms

Hurricanes

Major hurricanes

 (Category 3 or higher)

30-year average

12

6

3

NOAA

10 – 16

5 – 9

1 – 4

Colorado State Tropical Meteorology Project

14

6

2

The Weather Company

12

5

2

Colorado State University’s (CSU) Tropical Meteorology Project recently published an interactive tool that allows users to view and compare forecasts. Again, most groups have been predicting a season near or slightly above average 2018 hurricane season. But how accurate have these early predictions been in the past?

Early forecasting’s track record

Out of the past 16 hurricane seasons, less than half (seven, or about 44%) had a number of hurricanes within the bounds NOAA predicted in May.

NOAA Estimates and Actual Number of Hurricanes by Year

CSU predictions didn’t seem to fare much better – out of the last 23 years of outlooks, only one of the University’s hurricane occurrence predictions was correct. (To be fair, CSU only gives an integer prediction in their outlooks, not a range like NOAA).

And what about last year’s devastating storms? NOAA’s original 2017 outlook predicted an above-average season, with 11 to 17 named storms, 5 to 9 hurricanes, and 2 to 4 major hurricanes. During the catastrophic 2017 hurricane season, there were 17 named storms, 10 of which became hurricanes, and 6 of which were classified as major. The cumulative ACE (Accumulated Cyclone Energy, a measure of the overall duration and intensity of a season or individual tropical cyclone) for the season was nearly 225, compared to the May prediction of a “near-normal” 69 to 143. In other words, NOAA’s preliminary 2018 forecast is very similar to its 2017 outlook – which fell drastically short of predicting the severity of the record-breaking season.

Three of last season’s named hurricanes – Harvey, Maria, and Irma – fall among the top five costliest U.S. hurricanes on record, totaling $265 billion in damages. The human cost of the 2017 season was similarly devastating. A recent Harvard study estimated that Hurricane Maria caused 4,645 excess deaths in Puerto Rico alone, primarily due to interruption of healthcare services. Moreover, communities in Puerto Rico, the US Virgin Islands, and much of the Southeast US are still rebuilding infrastructure and recovering from last year, even as the new hurricane season begins.

Predict the average, prepare for the worst

Extreme weather events like last year’s storms are becoming the new normal, and we simply cannot accurately predict what a hurricane season will look like months in advance. As we approach future worsening storms and general climatic uncertainty, preparedness – regardless of forecasts – is our greatest defense against loss of lives and money.

While preparedness has improved in the years since Katrina, we are still just “good” at dealing with hurricanes (only on the mainland, and only after they strike). To get the best “bang for our buck” when it comes to disaster prep, we must invest in mitigation strategies, especially for our most vulnerable communities. Furthermore, our preparedness and mitigation work must be ongoing, not just as a part of recovery from previous events. This is particularly true in light of the uncertainty that comes with extreme weather events.

Determining return on investment (ROI) for preparedness is notoriously difficult. However, in recent years, increasingly costly storms are making the value of preparedness more evident to the public, and recent studies have gone a long way toward proving the adage that “an ounce of prevention is worth a pound of cure.”  Indeed, early this year the National Institute of Building Sciences showed that “mitigation funding can save the nation $6 in future disaster costs, for every $1 spent on hazard mitigation.” This means investment not only in preparedness, but also in resilience, is imperative.    

Investment in resilient infrastructure will prove particularly invaluable in the coming years. Healthcare depends on many critical infrastructure sectors to deliver services to patients. This means investment in healthcare alone is not enough –  we must evaluate our communities’ power grids, transportation systems, water systems, and more to improve their ability to withstand and respond to disasters. The deaths caused by interruption of medical care due to loss of telecommunications, power, and transportation in Puerto Rico after Maria demonstrate in stark detail just how important critical infrastructure is, especially after a disaster.

2018 hurricane season preparations

Following the 2017 hurricane season, Healthcare Ready is doubling down on its usual preparations and introducing new ones. Ahead of the peak of the 2018 season, we will be redesigning Rx Open, our free map that helps patients and emergency response professionals find open pharmacies in areas impacted by disaster. Later this month, we will release the results of our third annual national preparedness poll, which we use to help determine how public attitudes and actions regarding disasters may impact healthcare and policy. Finally, we will continue to work with our partners to identify lessons learned from our responses last year and while identifying the best ways to improve preparedness and resilience across sectors.

Even with the best science, hurricane season is inherently unpredictable. So while we plan for the coming season with forecasts in mind, it makes sense to hope for the average, but expect and prepare for a season as bad as or worse than last year.

Courtney Romolt

Courtney Romolt is a Program Analyst at Healthcare Ready, where she provides research and communications support for a wide variety of programs and initiatives. Before joining Healthcare Ready, Courtney worked with the World Resources Institute’s Initiative 20x20 to promote the business case for landscape restoration in Latin America and the Caribbean. Prior to this, she researched the environmental effects, economic trends, and regulatory framework of enhanced oil recovery as a Graduate Consultant for the NGO Clean Water Action. Courtney holds a Masters of Arts in International Economics and Energy, Resources and the Environment from the Johns Hopkins School of Advanced International Studies, where she also earned a Graduate Certificate in Chinese and American Studies in Nanjing, China. Courtney also holds undergraduate degrees in Integrative Biology and Global Studies from the University of Illinois at Urbana-Champaign.