Posted by: bmeverett | May 14, 2014

Another Bad Climate Study

With great fanfare, President Obama announced last week the release of the third National Climate Assessment (NCA), a periodic report required by law under the Global Change Research Act of 1990. The White House called the report “the most comprehensive scientific assessment ever generated of climate change and its impacts across every region of America and major sectors of the U.S. economy.” The climate chorus naturally echoed the refrain. The New York Times ran a front page headline, “U.S. Climate Has Already Changed, Study Finds, Citing Heat and Floods.” ABC News says, “Fed Report Says Climate Change ‘Has Moved Firmly Into the Present’”. If you Google “National Climate Assessment”, you’ll get 118,000,000 hits.

Far from being a fresh look at climate change, the report is just a tired rehash of all the old climate arguments. The Climate Community will love it, but nobody interested in climate change will learn anything new. For future reference, there should be three no-no’s for climate reports: (1) calling its conclusions definitive because it was written by scientists, (2) claiming that any meteorological event off the trend line is clear evidence of impending catastrophe and (3) presenting climate model simulations as evidence. If you eliminate these three problem areas from the NCA, there’s nothing left.

Take no-no (1), for example. The New York Times article starts out “The effects of human-induced climate change are being felt in every corner of the United States, scientists reported…” The implication here is that scientists are united in their view and that non-scientists lack standing to object. The White House tried hard to reinforce this view by explaining that the report had been prepared by 300 experts overseen by a 60-member Federal Advisory Committee. The report was subjected to extensive public review, etc.

Anyone who has participated in a study group understands the pitfalls. Somebody gets to decide who’s invited to participate and, more importantly, who’s not invited. Somebody gets to draft the report, and finally, somebody gets to decide which comments to incorporate and which to reject. Some studies are managed by people who try very hard to produce a high-quality report that advances our understanding of the issue. Others, however, are structured to produce a predetermined answer.

The study was chaired by Dr. Jerry Melillo, a Senior Scientist at the Marine Biological Laboratory in Woods Hole, Massachusetts and a Lead Author of the Intergovernmental Panel on Climate Change (IPCC). Dr. Melillo is a scientist of the highest caliber with distinguished credentials and an important position at a prestigious institution. But (and this is a big BUT), he has for years expressed a clear position on climate change. For example, Dr. Melillo was the co-author of the 2009 report “Global Climate Change Impacts in the United States.” At the press conference announcing the issuance of that report, Dr. Melillo stated, “It is clear that climate change is happening now. The observed climate changes we report are not opinions to be debated, they are facts to be reported.” Dr. Melillo has every right to state his view, but by doing so he has switched from scientist to political advocate. Scientists want to find the truth; political advocates want to win the debate. His appointment as Chair of the NCA determined from the outset what the report would say. To be clear, Dr. Melillo did nothing in any way unethical or unprofessional. He simply organized the work to show what he wanted to show. Does anyone believe that the NCA would have produced the same result under the chairmanship of Pat Michaels, Ross McKitrick or Stephen McIntyre?

Regarding no-no (2), the NCA’s portrayal of the current (and supposedly obvious) evidence of extreme weather events is an embarrassment. This problem persists in our climate change discussion for a couple of reasons. First, weather is a long-term phenomenon, and humans have very short data sets of personal experience. A bad winter may seem unusual, but who carries in his head enough data to determine how unusual it actually was? Second, extreme weather variability is the rule, not the exception.

DC snow stats

As an example, I’ve attached a chart showing annual snowfall in Washington, DC since 1888, when record-keeping started. Just a quick glance at the chart shows the enormous year-to-year variation and the difficulty of discerning a pattern. Snowfall in our nation’s capital has averaged 18.1 inches per winter, but the average annual variation from that average has been roughly 50%. There have been years when snowfall was nearly triple the average (2010 and 1899, for example) and years when there was virtually no snow at all (1973 and 1998). What are we to make of this?

All of the following statements are true.

1. 2010 was the highest winter snow accumulation in DC’s recorded history.
2. Three out of the five snowiest winters in the last 45 years occurred in the last 10 years.
3. Snowfall in the first 15 winters of the 21st century was 6% higher than in the last 15 winters of the 20th century.

However, the following statements are also true:

1. Three of the four snowiest decades in DC history were the 1890s, the 1900s and the 1910s.
2. DC snowfall over the last 50 years was 13% lower than snowfall in the prior 50 years.
3. So far, DC snowfall in the 21st century has been 14% less than snowfall in the 20th century, which was in turn 27% less than snowfall in the 19th century.

Since the data are all over the map, it’s easy to craft statements that support whatever position you wish. Extreme weather events occurred just as regularly a hundred years ago, before human greenhouse gas emissions became significant.

If I were to catalogue all the occurrences of this fallacy in the NCA, this post would be way too long, so let me give you one clear example. Page 41 of the report contains Key Message #8 “The intensity, frequency, and duration of North Atlantic hurricanes, as well as the frequency of the strongest (Category 4 and 5) hurricanes, have all increased since the early 1980s.” But we have data for Atlantic storms back to 1900, so why look at just the last 30 years rather than the whole data set? Furthermore, climate change is a global phenomenon, so why not look at hurricanes on a worldwide basis?

Worldwide, there were 318 hurricanes in the 1980s, 347 in the 1990s, 308 in the 2000s, and we are on track for 275 in the 2010s. See a pattern? I don’t.

There were 10 Category 4/5 North Atlantic storms in the 1980s, 14 in the 1990s and 23 in the 2000s, so the NCA’s statement about major storms is true, but I could say with equal validity that there have been no Category 5 storms in the last 6 years compared to 8 in the previous 6 years. Does this mean that severe storms are decreasing? If the high severe hurricane activity in the 2000s was in fact caused by human-induced climate change, what caused the high activity levels in the 1930s (16 Category 4/5 hurricanes) when atmospheric carbon dioxide levels were much lower than they were today? What caused the drop in the 1970s (8 Category 4/5 storms) when atmospheric carbon levels were increasing?

In many ways, no-no (3) is the most troubling. The NCA frequently conflates statements about what is happening now with statements about what is predicted to happen in the future. Chapters 16-25 address regional impacts. The overall regional section starts out with the statement that “One common challenge facing every U.S. region is a new and dynamic set of realities resulting from our changing climate. The evidence can be found in every region, and impacts are visible in every state.” This theme was repeated extensively in the press. For example, US News and World Report states “[The NCA’s] main conclusion is that climate change is already here. Worse, its effects have arrived with a speed and severity that few thought possible.” Each regional chapter includes a series of “Key Messages”. Have a look at the Key Messages for the Northeast from page 372:

1. Heat waves, coastal flooding, and river flooding will pose a growing challenge to the region’s environmental, social, and economic systems.
2. Infrastructure will be increasingly compromised by climate-related hazards, including sea level rise, coastal flooding, and intense precipitation events.
3. Agriculture, fisheries, and ecosystems will be increasingly compromised over the next century by climate change impacts.
4. While a majority of states and a rapidly growing number of municipalities have begun to incorporate the risk of climate change into their planning activities, implementation of adaptation measures is still at early stages.

All of these Key Messages are predictions of dire consequences in the future, not conditions that have already happened. The text includes some disjointed observations of actual climate events, but these observations are mixed in with predictions. The same is true for the Key Messages from the other regions.

Predictions are not evidence of anything, unless they are based on models which have a meaningful theoretical and empirical basis and a track record of success. We can predict with real accuracy what would happen if we drop a brick from the top of the Empire State Building because we understand not only how gravity works, but also the more subtle impacts of air pressure and wind. Climate models reflect our limited understanding of climate dynamics. These models will become a useful basis for policy when and only when they are able to make correct predictions. Furthermore, these predictions must be falsifiable. It’s meaningless to say that our model predicts extreme weather events, since we always have extreme weather events. The models would have to predict which extreme events would occur and where. So far, they can’t predict anything.

Nice try, but nothing new here.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: