Earth Sciences: The Study of Superfloods

Victor R. Baker

Large, high-energy floods are both rare and dangerous. Evidence of their impacts in the geologic record is often subtle, but the greatest obstacles to advancing the knowledge of superfloods have come from misapplied scientific logic. Particularly troublesome are flawed notions of hypothesis testing, verification, and the principle of simplicity. Contrary to conventional views of scientific methodology, there has never been a general theory of superfloods that could be tested, confirmed, or falsified by observation and experiment. Instead, as in much of geology, observation has preceded theory, and understanding has emerged as previously unrecognized phenomena were discovered.

Methodological problems with the study of superfloods began early, at the inception of geology as a science. The influential 19th-century geologist Sir Charles Lyell passionately advocated a regulative principle for validating the inferences that geologists make about the past. Presuming that geologists reason by induction, Lyell thought that such a principle was necessary if geological inferences about past causative processes were to achieve the same kind of certainty as those made in experimental sciences like chemistry and physics. It was Lyell’s most capable critic, the Cambridge polymath William Whewell, who in 1832 named this principle „uniformitarianism.“ Epistemological uniformitarianism holds that scientifically reliable inferences about the past must be confined to invoking only the slow-acting, low-magnitude processes currently in evidence to human observers.

This principle was applied blindly by Lyell’s intellectual descendents well into the 20th century. In the 1920s, Bretz’s documentation of the spectacular effects of late-glacial flooding in the Channeled Scabland region in Washington state met with immense criticism from the scientific community. Not until the 1960s was it generally accepted that this flooding was caused by catastrophic failure of the immense ice-dammed Glacial Lake Missoula along the southern margin of the Cordilleran Ice Sheet, which covered the northwestern mountains of North America about 20,000 years ago.

Over the past 40 years, evidence has accumulated for catastrophic failures of ice-dammed lakes, overflows of lakes that had formed along ice margins, and subglacial outburst flooding in the many river systems associated with the immense continental ice sheets of the last ice age. The southern margin of the Laurentide Ice Sheet that covered northeastern and north-central North America was episodically drained by outburst floods  These freshwater discharges greatly influenced ocean circulation and climate at the end of the last ice age.

Even more extensive evidence of superflooding has been documented in Asia (…)

(…) These highly controversial studies of superfloods show that flood science has not achieved the universally accepted valid scientific methodology[[ envisioned by Lyell. Instead, it is my view that superflood studies are motivated by a notion introduced by Whewell, who proposed in 1840 that productive scientific hypotheses work toward achieving „consilience,“ a kind of confirmation through the unexpected connections and explanatory surprises they engender.

complete article www.science.com (autorisation required)
or in Science: Volume 295, Number 5564, Issue of 29 Mar 2002, pp. 2379-2380