Edited Version June 2, 1999 Transcript
EIIP Virtual Library Online Presentation

"Hurricane Season 1999"

Roger Pielke, Jr.
Environmental and Societal Impacts Group
National Center for Atmospheric Research
Boulder, Colorado

EIIP Moderator: Amy Sebring

The original unedited transcript of the June 2, 1999 online Virtual Library presentation is available in EIIP Virtual Forum Archives (http://www.emforum.org). The following version of the transcript has been edited for easier reading and comprehension. Typos were corrected, date/time/names attributed by the software to each input were deleted but the content of questions and responses are as stated by each participant. Answers from the participants to questions by the audience are grouped beneath the appropriate question to facilitate meaning.


Amy Sebring: Welcome to the EIIP Virtual Library!

• For the benefit of our first-timers, when you see a blue Web address, you can click on it and the referenced Web page should appear in a browser window.

• After the first one, the browser window may not automatically come to the top, so you may need to bring it forward by clicking on a button at the status bar at the bottom of your screen.

• Please do not send private messages to the moderator or our guest during the session as it makes it difficult for us to follow the discussion.

• Right before we begin the Q&A portion we will review how to submit questions/comments.

• For background info on today's session, please see <http://www.emforum.org/vlibrary/990602.htm>


We are pleased to welcome Roger Pielke, Jr. from the Environmental and Societal Impacts Group of the National Center for Atmospheric Research in Boulder, Colorado.

Roger has conducted research into the impacts of hurricanes, and has uncovered some facts that may surprise you. He, with his father, have written a book titled appropriately, Hurricanes: Their Nature and Impact on Society. Since we just began the official start of hurricane season yesterday, we are very glad to have him with us today to discuss the upcoming season.

Welcome Roger, and thank you for being with us.


Roger Pielke: Hello everyone and thanks for coming! This is my first "Internet talk," so I hope it goes well. Thanks to Amy Sebring, I'm sure that it will.

Today I am going to present some of the research on hurricanes that we, along with colleagues, have been doing here at the Environmental and Societal Impacts Group (ESIG) at the National Center for Atmospheric Research (NCAR). But first, I thought that I'd spend a minute introducing myself and also introducing you to ESIG, as many of you may not be familiar with us.

I am interested in how science is connected to decision-making. Here at NCAR, I study extreme weather impacts and responses including the use and value of forecasts, policy for global climate change, and the management of science by government.

I have a home page at http://www.dir.ucar.edu/esig/HP_roger.html. If you load that right now you can put a face to these words! On my home page you'll see a list of publications any if which I would be happy to share with you, just send me an email with your request <[email protected]>.

At NCAR, ESIG is a small group of mainly social scientists who study the interactions of the atmosphere and society. So in addition to natural hazards, we study things like climate change, climate variability, the use and value of forecasts, and various policy issues that are related.

You can learn more about what we do from our WWW site at http://www.dir.ucar.edu/esig and you can browse that later. Now on to today's topic --- Hurricane Season 1999!

For the remainder of the session I am going to present some information on (1) Hurricane Impacts, (2) Hurricane Forecasts, and (3) Hurricane Policies. I intend to keep my commentary rather brief, so that we will have some time for conversation. In addition to presenting information, I hope to be able to learn from you as well!

Hurricane Impacts

How bad is the threat of hurricane impacts in the United States? This is a question that underlies the intensity (read $$$$) with which policy makers act to prepare for hurricane impacts. In the 1990s many policy makers, in government, insurance, etc. have become aware of the fact that hurricane losses are increasing dramatically. Have a quick look at http://www.dir.ucar.edu/esig/HP_roger/hurr_norm/rpbook1.gif.

Some people have ascribed the increase in losses to more intense or more frequent storms, such as Newsweek magazine, which ran a cover story in 1996 saying "blame global warming" for hurricane impacts. The US Congress even issued a report on Natural Disasters in 1994 linking the increase in hurricane losses to changes in hurricane climatology.

Unfortunately, these ideas about causality were simply wrong. In fact, during the period when hurricane impacts have increased dramatically, the frequency of the most intense storms, i.e., Saffir-Simpson Category 3-4-5, has actually been depressed from previous periods!

So when Newsweek and the US Congress "blamed" the climate for the recent increase in impacts, they misinterpreted what was actually going on. If you'd like more information on hurricane climatology you can see the work of Chris Landsea, with whom I have collaborated with on work on impacts. http://www.aoml.noaa.gov/hrd/Landsea/landsea_bio.html

Instead of climate, it is society that is making changes that lead to more damages. Motivated by the frequent misinterpretations of the historical record, Chris Landsea and I published a simple study in 1998 that asked the question: If storms of the past made landfall today, what sorts of damages should we expect?

To do the analysis, we adjusted past losses for inflation, population, and wealth in the counties affected by the storms. For later reading the full paper can be found in HTML and PDF format at:


We found that the rapid increase in damages was reflective of the massive influx of population to exposed coastal regions. A graph of the adjusted damages can be found at http://www.dir.ucar.edu/esig/HP_roger/hurr_norm/rpclhur.gif

Let me provide some information from the graph.

Hurricane Andrew, at $30 billion in losses in 1992, was bad, but not nearly as bad as the estimated $75 billion in losses from the 1926 Great Miami Hurricane. And consider that the state of Florida alone saw losses of about $50 billion from 1944-1950 --- the same total as from 1951 to the present! And every year more people and wealth is placed in exposed locations. Data like this, as well as events like Hugo, Andrew, and even Opal and Fran has elevated interest in better understanding the threat of hurricanes in the US.

So far I have talked only about economic damages. Human losses are important as well. Recently there was a news article discussing whether or not "Mitch-like" losses might be possible in the US. See


I'm not sure that losses exactly like Mitch can occur in the US, but there is no doubt that a large loss of life is possible. This turns us to our next topic, hurricane forecasts.

Hurricane Forecasts

Hurricane forecasts take place on two distinctly different time scales. One, which I won't say much about, is the seasonal forecast, pioneered by Prof. William Gray and his team at Colorado State University, if you are interested see http://tropical.atmos.colostate.edu/forecasts/index.html. An updated forecast from Prof. Gray is due this week, we can talk more about these in the discussion if you'd like.

I'd like to turn to an apparent paradox in the area of hurricane landfall forecasts and related emergency management warnings. I am hoping that you will help me to better understand this paradox.

Scientists have noted in recent decades improvements in their forecasts of hurricane landfalls, with improvements defined as reducing the average difference between where the storm is forecasted to go and where it actually goes. In other words, hurricane track forecasts have become more accurate, by about 1% annually.

At the same time it has been noted that the average miles-of-coastline warned per storm has INCREASED during this period. This means that as hurricane forecasts have improved, the number of people warned per storm has increased, (called "overwarning") thus the paradox.

There are of course possible explanations why this apparent paradox exists, including:

a) the desire of emergency managers (and elected officials) to base evacuation decisions on the NHC warnings, and thus request warnings be extended to cover their communities

(b) a desire throughout the evacuation decision process to avoid the error of a strike on an unwarned population, (thus, the forecast improvement could have translated into lower risk) and

(c) the fact that more and more people inhabit the coast, meaning that evacuation times are much larger, making necessary longer lead times and greater lengths of coastline warned.

Unfortunately, in spite of the existence of these hypotheses, it has not been convincingly demonstrated why the coastline-warned-per-storm has increased during a period of decreasing forecast errors.


Given the large costs involved with overwarning, both in unnecessary preparations and in potential public response, it would seem to be in the best interests of forecasters, policy officials, and the general public to obtain a greater understanding the use of hurricane forecasts. So the bottom line would seem to be that we, as a community and a society, have invested a great deal to improve hurricane forecasts and to document that improvement, but have not necessarily followed through to make sure that those improvements are effectively or systematically utilized.

As a result, it might be that we are not benefiting as much as we could be from the spectacular advances in the science of weather. In the discussion I'd appreciate your input on this subject. Since I have gone on for so long, for the last section, (3) Hurricane policies, I'd like to simply raise a few issues for discussion.

Hurricane Policies

Insurance. Who should pay for hurricane losses? Should public money underwrite private insurance? Should rates be charged at levels that are actuarially sound but perhaps out of reach of many? These are difficult questions that seem to come up whenever hurricanes and insurance are discusses together. There are no easy answers.

Federal disaster declarations. In theory the Federal role in disaster response is clear. But in practice the availability of Federal support might in some cases serve as a deterrent to mitigation. Also, the variation in different Administration's approach to disasters could serve to limit mitigation. What is the proper government role? How can we balance the need for a "safety net" with the perverse incentives of post-disaster aid? Again, no easy answers.


In closing, I'd like to place US hurricane losses in perspective. Yes, they are significant and yes, they matter a great deal, particularly to those directly affected. But when we place such impacts into a broader global perspective, and compare with events in Central America and in the Indian region, we quickly see how fortunate we are and how well we have in fact prepared for the events that we do face.

Thanks are due to the many people who have dedicated their lives to improving our societal responses to hurricane and other extreme events. I hope that in the future we can effectively transfer some of that knowledge and capacity to places whose responses don't match the success that we have enjoyed. Thanks for logging in!

Amy Sebring: Thank you, Roger. If you have a question or comment for our speaker, please input a question mark (?) and wait for recognition from the moderator before sending your question. You may compose your question and have it ready to submit but do not send it in until you are recognized. First question, please.

[Audience Questions]


Audra Kunf: Rgarding this 'paradox'. Seems, from my perspective, that it occurs mainly due to that perverse incentive of post disaster aid. Is it possible, that in hurricanes similar to EQ's, that government tends to over-dramatize to maximize potential for Federal Dollars?

Roger Pielke: There are of course highly political factors involved in the post disaster declaration (PDD) process. But as far as the paradox, this has much more to do with evacuation strategies, which also has politics involved. So the point I'd like to raise is that if we are spending money to improve forecasts, we should also invest a bit to ensure that they are effectively used.


Amy Sebring: Roger, I have a question regarding funding for hurricane research. My understanding is there is not parity between hurricane hazard and earthquake hazard. Is this true and if so, why is that?

Roger Pielke: This is a point often raised by hurricane researchers and dismissed by earthquake researchers. It is true that the Federal government spends more on EQ research than hurricane. It raises the question of how we should set our priorities for science. For instance, if you want to save human lives with respect to hurricanes, then put some money towards capacity building in Latin America and the Caribbean, not more research. And of course, the EQ "lobby" has been stronger in the past than the hurricane "lobby".

Part of the reason that we did the impacts study was to provide policymakers with a more rational basis for deciding the importance of hurricanes with respect to other priorities.


Avagene Moore: Comment on insurance. I realize the cost would go out of sight if premiums were based on real costs to people living on the coast, however, it might stop building in high-risk areas. Frankly, why should we underwrite that cost?

Roger Pielke: Avagene- you have hit upon a raging topic of debate.

Avagene Moore: Politically unpopular in some circles.

Roger Pielke: The private insurance industry does not insure for floods, but they do for hurricanes. If only those people exposed to hurricanes paid premiums, in many places the costs would be unaffordable, and probably fall back to the public. But the purpose of insurance is to share risk, so people in one place balance risks with people in another. What society has to be careful about is the unintended consequences of instruments like insurance and public assistance, which can actually increase vulnerability.


Amy Sebring: Roger, can you comment on the apparent importance of code enforcement in reducing impacts?

Roger Pielke: After Andrew, the Miami Herald conducted a block by block study of the damage. They found that damage occurred in a wide range of places well outside the area of maximum winds. Some of this can be explained by small scale winds, but the main explanation was that in spite of the South Florida Building Code, the strongest in the nation, many houses did not meet the code. This was attributed to poor building practices, lack of effective enforcement, and even a change in the style of homes that were built. It points to the fact that we need more than good laws, we also need to follow through and ensure that they are meaningful.


Amy Sebring: Anyone else with a question or comment for Roger? He is looking for some input on "overwarning".

Christopher_Effgen: Actually a few, but I am not sure I am ready to ask them. Roger has an interesting perspective that is new to me and I like to ask considered questions.

Amy Sebring: Well start with one. Pop it in when you are ready.


Audra Kunf: How much of a problem is 'overwarning'? I mean--if it's an 'all the time thing'--could be like crying wolf. But, if there's a major event --- isn't erring on the side of caution appreciated?

Roger Pielke: Certainly, but it is the definition of "caution" that matters. It has been estimated that it costs upwards of $1 million per coastal mile when an evacuation is ordered. For instance, when Georges approached New Orleans, but did not hit, it cost an estimated $50 million. This is serious money, and someone has to pay.

So there are pressures upon the forecasters to be as accurate as possible. At the same time, if the forecast is blown, or someone makes a bad decision, jobs, embarrassment, etc. are all at stake. Thus, there is a complex calculus involved in how to translate meteorological forecasts into practical advice for coastal communities. And my view is that the forecasting community is just now beginning to deal with a lot of those complexities.


Christopher Effgen: You seem to indicate that hurricane probability warnings are in some way subject to outside the weather service influence?

Roger Pielke: Christopher --- yes, that is what I am saying!


Christopher Effgen: You seem to have a position on some of the issues that mirrors those of the insurance industry and a position that seems to advocate some sort of federal involvement in building code enforcement?

Roger Pielke: I am a big advocate of LOCAL building code enforcement. The map that I describe before by the Miami Herald showing damages could have been produced before Andrew! We don't need a hurricane to show where we are vulnerable. And I'm not too sure about mirroring the insurance industry, as that is a varied community with many views, some of which I am sure that agree and some that would disagree with what I have presented today!


Amy Sebring: Re: Christopher's question, my experience is that the probabilities are calculated from models, but that the watches and warnings issued are more subjective even though accuracy is improving relatively. We still have a ways to go towards absolute accuracy, do we not? Is there a new research project about hurricane behavior as it approaches land especially?

Roger Pielke: Yes, good point Amy. There is a distinction between hurricane forecasts and hurricane warnings, they are not the same thing. Warnings are the societal manifestation, or outcome, of the forecast process. There is a research program called the US Weather Research Program that is focusing on 'hurricanes at landfall". It seeks to improve both forecasts AND our understanding of how those forecasts are connected to warnings and other decisions. <http://uswrp.mmm.ucar.edu/>.


Russell Coile: Didn't the Miami Herald find that a contractor had bribed a building inspector to not inspect the shoddy roof construction on hundreds of houses? And that Habitat for Humanity was proud that all of its houses built with TLC had no damage.

Roger Pielke: The Miami Herald found A LOT of "dirt", like inspectors who inspected houses at a rate of 1 every 6 minutes including driving time. And there was a complex pattern of influence between overseers, builders, and politicians that did not help the process.


Christopher Effgen: As for mapping damage probabilities before hurricanes, the same can be done before earthquakes, and I think that both projects could be undertaken to the public benefit.

Roger Pielke: I agree. And one does not even have to invoke probabilities, one could simply map buildings with respect to the existing level of the code.


Avagene Moore: Roger, what type of comparisons have you made between hurricane impacts in the Caribbean for example, vs. the US? I am interested because of where I am at the moment.

Roger Pielke: Along with Chris Landsea, a scientist from Cuba name, Jose Rubiera, we are working on a project that seeks in a similar manner to address the question, "what would the impacts be of past storms should they hit today". Like the US, we have found that economic damages are in fact lower than we might expect in more active periods. And most troubling is the finding that Mitch was not at all a rare occurrence. We should probably expect more such disasters unless we take action.


Audra Kunf: Just a comment re: EQ mapping pre-event --- that IS being done in California. But it's slow going convincing the government and private sector that it can minimize impacts. Any activity pre-event costs $$$ and we've nurtured a society that expects to be taken care of AFTER the fact; so why spend monies before something happens?!

Roger Pielke: Exactly. The best time to prepare for a disaster is right AFTER a disaster when a "window of opportunity" opens and people are more receptive to change. The challenge is to maximize the impact of such a window, having a well thought out plan well in advance is one way to take advantage of the inevitable events that will occur.


Amy Sebring: My concern about overwarning is a traffic disaster. Too many cars entering limited highways all at the same time. From Director Jarrell's remarks in WeatherZine last month, it looks like he may share this concern? Do you, as well?

Roger Pielke: Yes. The hurricane community received a shock in 1995 when Hurricane Opal rapidly intensified to Cat 5 strength as thousands of people were stuck in traffic in Alabama. I have had more than one forecaster tell me that if Opal had remained that strong (it weakened considerably) that they feared 100's of deaths.


Audra Kunf: But strengthening AFTER an event has its own problems. FEMA will not 'improve'. Stafford Act funds only apply to pre-event restoration. Maybe we have become too dependent on Federal funding to ever be truly 'self-sufficient' for future preparation?

Roger Pielke: I agree. This is one example of the "unintended consequences" that I mentioned earlier. In theory, 'no improvement" makes sense, perhaps to avoid exploitation. But at the same time, it can simply recreate the conditions that led to disaster in the first place!

Avagene Moore: To add to what Audra said. That is exactly the same mindset or attitude about mitigation that is being reported by all the countries in Central America and the Caribbean. I have been struck by the fact that we all face similar habits and the same obstacles to prevent repeated disasters. Also no money and few resources. A few good people who carry the banner for preparedness and mitigation.

Amy Sebring: This is all we have time for, but please stick around afterward for some further open discussion.

Again, our thanks for sharing this information with us, Roger. A text transcript will be up later today, and a reformatted online version up early next week.

Roger Pielke: Thanks all!

Amy Sebring: Speaking of next week, Kellye Junchaya will be hosting the Round Table next Tuesday, (with any luck on connections) from the Community and Family Preparedness Conference at Mount Weather.

Next Wednesday, we are pleased to present Carole Macko from the EPA's Chemical Emergency Preparedness and Prevention Office. You may realize that later this month is the deadline for regulated businesses to submit their Risk Management Plans to the EPA. Carole will discuss what happens after that.

Thank you all for joining us today. We will close down the Library for now, but you are invited to adjourn back to the Virtual Forum room for some open discussion and to express our appreciation to Roger.