27 November 2012

The Costs of Natural Disasters

I've posted previously on the contributions of FEMA to natural disaster response in the US. In general, there are four recognized stages of emergency management that also translate well to an understanding of disaster-related spending:
  1. Preparedness, involving immediate planning for anticipated disasters because of something in the forecast or because it's a slow-building event. Communities may order evacuations, residents may be asked to stock up on food and seek shelter, and supplies and personnel may be positioned to help in the later response process. 
  2. Response, with attention to the affected population's immediate and short-term needs such as food, water, shelter, and transportation. This is usually a FEMA activity, bringing the most obvious and direct influence of federal preparation and infusion of federal funding.
  3. Recovery, the long-term process of rebuilding to the pre-disaster levels of economic activity. This is usually a post-FEMA activity, when communities and residents have had the chance to evaluate their losses and make insurance claims, business owners have the chance to re-open their doors, and opportunistic decisions are made to move or close or even expand enterprises.
  4. Mitigation, the effort to restore protective measures to pre-disaster levels, and possibly to upgrade that protection in advance of the next event, with a goal of reducing or preventing damage in future events, sometimes under the term "disaster risk reduction." This may actually be the most complex component of disaster-related spending, taking great effort to keep decision-makers' attention on such priorities at times outside of the disaster itself.
The cycle comes back around to #1 when a disaster seems imminent. Some accounts of the cycle like to start with #4, as if people are willing to work on long-term mitigation of disasters that they've never seen or experienced before. However, that's not what seems to happen in reality, based on what I've seen, read, and experienced in my own lifetime. People who move to a new city or home seem to need first-hand experience of at least one disaster of moderate proportions before they get educated on the dangers of their new location and make an effort to plan for the future. I don't have any research to back this up, but I would suggest that people who survive one moderate disaster are more likely to survive subsequent, even larger events because of what they learned form that initial experience. Of course, living in a community that is tightly-knit overall, and well-coordinated when the time for disaster preparedness comes around, will certainly help lessen the impact of that first disaster experience.

Different types of disasters have characteristics that provide communities the opportunity to seek different levels of protection. For example, climatic records tell us the largest storms that affect a particular area, so there is guidance for engineers to develop structural protection measures such as levees and seawalls. Climatic records also provide a history of dry seasons and drought, informing planners and farmers of the need to prepare water supplies for those periods. Historical records of the residents in a location can help communities recognize the likely impact of a disaster event, and that information can guide their preparation and their anticipation of the response needs. It would be great if every community had the cash to prepare for and protect against every type of disaster, but that doesn't happen, so there is a consistent need for response and recovery spending after disaster events. Hopefully, there is also a chance for learning and re-evaluation of priorities, but that rarely happens in the response phase itself, so we can only hope that the the memory of a diaster lingers just long enough to teach its lessons. Unfortunately, that memory usually comes at the cost of lives.

Some events cannot be forecast yet: earthquakes strike suddenly, but there are some advance signals that scientists are studying constantly, and communities like those in Japan and California have worked with engineers to develop "quake-proof" building codes and practices. That level of preparation, based on experience and experiment, is aimed at minimizing the loss of life in future events and reducing the cost of response and recovery after the event has passed. These are the types of events, with little or no warning, that take the largest toll in terms of human lives. In general, the level of warning translates directly to the number of lives spared. Tornado warnings with just a few minutes' notice in many parts of the US, through National Weather Service broadcasts and local alarms, have saved countless lives over the past several decades.

But many disaster events can be forecast to some degree. Storms that produce those tornadoes, among other wind- and flood-related impacts, can be tracked on weather radar and forecast with some skill. Hurricanes move onshore after some time over the ocean where they are watched and analyzed from numerous perspectives. We know that certain communities will be most affected, and the weather forecasts help us determine which communities are under threat and then which will likely need the most attention afterward. Presidential emergency declarations prior to or during a storm event, and then disaster declarations after the event, help prioritize the immediate preparations and then post-event response and recovery efforts, and channel funds in the directions of greatest apparent need. Even with the best available science and engineering, we humans continue to build and live in vulnerable areas, so disasters still exact huge costs in the affected communities.

Several sources have attempted to demonstrate and draw attention to the rising costs of natural disasters,  both in the US and around the world, and especially in recent decades as reporting of such impacts has improved over time. The ultimate cost, in terms of human lives, was tallied by the US National Weather Service as shown by the Washington Post Wonkblog:

From the Washington Post Wonkblog, posted 30 October 2012 at this location.

Using a report from Munich Re (a global reinsurance company), the journal Nature counted the overall number of disasters reported around the world over the past three decades:

From Nature, posted 10 January 2012 at this location.

The Economist has also attempted to tackle the story of rising disaster costs on at least two occasions recently, both relying on Munich Re reports:

From The Economist, posted 21 March 2011 at this location.
From The Economist, posted 14 January 2012 at this location.

That latter Economist article is an excellent review on the macro-economics of natural disasters, and it was quite satisfying to see that level of discussion on these issues in such a highly-regarded international news magazine. It should be noted, however, that these numbers address insured losses, and that additional uninsured losses are some unknown amount greater in almost all cases.

Back to the US and even more up-to-date, Bloomberg Businessweek recently produced this interesting graphic showing the comparative costs of disasters in the US over the past two decades, including an early estimate of response and recovery costs following Hurricane Sandy:

From Bloomberg Businessweek, posted 1 November 2012 at this location.

In a number of sources, it is clear that some researchers and reporters are attempting to make apparent any potential influence of climate change on the rising numbers and costs of natural disasters. However, as linked in the Nature article that I referenced above, the careful science of such attribution is still in the early stages of development. It is already easy and plain to say that some of these events would not have been disasters if not for humans pressing on their natural environments, building levees and towns where the river wants to run or cities in coastal flood zones with inadequate seawall protection, but that doesn't mean that human activity affected the power of the storm itself. I do believe that someday we will be able to say with full certainty "yes, anthropogenic climate change made this storm/drought/flood worse, and it would not have been so costly if we had been more responsible with our greenhouse gas emissions." However, the method of proving that is very strict, and it takes time to compile the evidence. I often think that neither scientists nor journalists nor the public really understand that this language scientists use to convey uncertainty, words like "possibly" and "likely," mean different things to different people. The IPCC has their definitions spelled out clearly in various reports and a Glossary (pdf) (see the definitions for "confidence," "likelihood," and "uncertainty"), and that's what I prefer. However, until everyone using and reading those terms comes to a common understanding of their definitions, I also recognize that those terms need to be defined unambiguously. Yes, the trends of climate change and natural disasters are tracking upward together, suggesting a correlation that calls for further investigation, but the day that we get to "very high confidence" or "very likely" (both meaning >90% probability) is still some time away. The basic uncertainty in making such a claim has many components still to be resolved, and there's still that 10% of leeway in the probability that gives some people justification for their continuing doubt.

Climate scientist James Lawrence Powell just posted a meta-analysis with graphics showing that, in the period 1991-2012, almost 14,000 professional, peer-reviewed journal articles on climate with nearly 34,000 individual authors were published supporting a scientific consensus on anthropogenic climate change. He could find only 24 peer-reviewed, published articles that dissented, and those were poorly-cited articles at that. Yet we know from media reports and journalistic language ("Scientists say ...") that there are still large parts of the US and global populations that deny climate change even exists, let alone that human activity might be causing at least some of that change! The process of science in consensus building is amazing, but it takes time and a willingness to test and re-test hypotheses and to accept observations as evidence. The IPCC 5th Assessment Report (AR5) due for release in 2013 will likely set a new standard regarding the actual certainty of anthropogenic climate change, and every scientific study published thereafter will raise or (unlikely) lower that certainty by fractions of a percentage point. My own sense, from the literature and evidence, is that human activity is "very likely" (>90% probability) causing observed climate change, and that the 2-degrees-C mark is "virtually certain" (>99% probability) of being reached before the end of the current century, or likely even earlier. However, I'm not specifically trained as a climate scientist, so I'll leave those determinations to the experts. I do attempt to specialize in the water-related impacts of such changes that show up in various parts of an exceedingly complex system, both now an in the future, and the fundamental science of climate change suggests that we have already done (and continue to do) dangerous and degrading things to the planet. Society doesn't need to wait on scientific journal articles and subsequent news stories to plan their response to events that are already occurring, to become better prepared for similar or bigger events in the future, and to reform the ways that communities consistently build and re-build in harm's way. The goal of community planning, and especially the mitigation phase of the disaster cycle, should be resilience in the event of disaster. That is a term and idea on which I have wanted to write for some time now. It is part of the discussion now, near the end of the response phase and at the beginning of the recovery/rebuilding phase in the northeastern US following Hurricane Sandy, so this is as good a time as any to talk about it here.

Any search on Google Images for "natural disaster infographic" (and variants on those terms) produces hundreds of results, so these are just some examples of the wealth of information out there. However, as with any search for infographics based on real data sources, it's important to check for references. The best infographics provide, right on the image, their references (and sometimes web links) for the data used in their graphical design. I have attempted to use only infographics that provide such information, or at least link to their sources from a web page that is obviously connected to the infographic. Where I have used figures from other blogs and media and journal articles, I have attempted to be sure to include all reference and link information. As always, if you find that a source is missing or a link from this blog is broken, let me know so that I can resolve the issue.

No comments: