Norman Fine's comments on Stephen Ritter's well-done article on global warming should not go unchallenged (C&EN, Dec. 21, 2009, page 11; Feb. 15, page 8). When considering uncertainty, we want to know the statistical probability that a given result is part of a specified population or sample thereof. If we examine the probabilities associated with values for the unit normal deviate (standard deviation), using either a t-distribution or a z-score-versus-probability table, we see that a 90% probability corresponds to 1.645 standard deviations, and 95% probability corresponds to 1.960 standard deviations.
Thus, it is incorrect to say, as Fine does, that the difference between 90% and 95% confidence is one standard deviation. It is, in fact, 0.315 standard deviations, a much smaller number.
The 95% confidence level is often thought of as a gold standard in assessing statistical significance, but it is not universally applicable. Other levels are often used in complex situations, and 90% is fairly commonly employed. Only if you were down at the 50% probability level for a two-valued hypothesis would you be justified in merely flipping an unbiased coin to decide. Risk should be viewed as the interaction of probability of occurrence and severity of the consequences, and addressed accordingly.
William H. Flank
Recent letters questioning climate change continue the usual distortion of fact and concept by skeptics (C&EN, Feb.15, page 4, and online). One letter claims that the Northwest Passage was as frozen in 2009 as it was during Roald Amundsen's 1903–05 journey. Another implicitly equates the harrowing 1940–42 passage of an ice-fortified schooner with "opening of the Northwest Passage in 1940." These familiar distortions are exposed by examination of NASA images and historical accounts (see earthobservatory.nasa.gov/IOTD/view.php?id=40046 and en.wikipedia.org/wiki/Northwest_Passage).
The political strength behind these distortions is frightening. At the recent "Great Climate Change Debate" at Rice University between Richard Lindzen of MIT and mainstream climatologist Gerald North of Texas A&M, I was stunned by the broad anger in the audience toward the scientists involved in "Climategate." Like those scientists, I also strive as a reviewer to keep junk science out of the literature and use "tricks" to reconcile computer modeling with experimental results. I avoid any discussion of science outside of work.
Where does this populist anger originate? In prosperous times, people were preoccupied with work and family, but a single-minded ideology of maximizing corporate profits has gutted domestic manufacturing and threatens the middle-class hopes of younger workers. Elimination of professional jobs is now creating waves of unemployed scientists. Anyone with limited job security naturally worries, "I may be next."
Meanwhile, entrenched academic and government scientists believe they are entitled to handsome salaries and first-class research facilities. Is their taxpayer-funded research truly building a foundation for increased private-sector employment or improving our health, or are rising government research budgets simply worsening the federal deficit and creating more unaffordable regulatory burdens (C&EN, Feb. 15, page 16)? Resentment against the arrogance and privilege of the scientific establishment has found its perfect target in climatology.
Our civilization is facing the consequences of a devastated industrial base, a demoralized workforce, a failed economic ideology, paralyzed government, unsustainable entitlements, and climate change. In light of this colossal failure of liberal democracy, the skeptics' letters become meaningful. Half of us are wailing while the rest are rearranging the deck chairs. United we can confront our failure or collectively perish.
William K. Wilson