(Bloomberg) — When Hurricane Harvey dropped 60 inches of rain onHouston in August, some described the storm as "biblical."One of America's leading hurricane scientists has now sharpenedthat assessment. 

Houston rainfall

"By the standards of the average climate during 1981-2000,"MIT's Kerry Emanuel writes in Proceedings of the NationalAcademy of Sciences, "Harvey's rainfall in Houston was'biblical' in the sense that it likely occurred around once sincethe Old Testament was written."

Related: Tallying Harvey's tragic price, one storm-ravagedhome at a time

Emanuel's analysis attempts to answer a series of questionscritical to recovery and rebuilding in Houston andelsewhere: "Should buildings, homes, roads, and associatedinfrastructure be built in the same place again?" he wrote. Arebuilding codes, levees and sea walls tough enough for thefuture? As the world warms, every community will have tograpple with these questions on their own.

Hotter seas, more humid air

Climate scientists continue to project that as the century goeson, hotter seas and more humid air are likely to make tropicalstorms more intense. The first volume ofthe latest U.S.National Climate Assessment, released by the Trumpadministration earlier this month, reports scientistshaving higher confidence in their finding that storms willcarry more precipitation.

Global warming has already helped increase the annuallikelihood of Harvey-like rainfall sixfold since the end ofthe 20th century, to about 6%, Emanuel writes. Beforethis century's out, that probability may rise to18%. 

Researchers such as Emanuel are working with hurricane recordsthat are historically short and incomplete. The best data have comesince 1980, and while scientists have greater confidence in theirprojections going forward, the historical record is stillvery noisy.

Powerful computer models

Information can be gleaned by combining observations withwell-understood meteorological physics and with powerfulcomputer models that can simulate past trends and "run"thousands of future storms. 

"It's a ground-breaking paper," said Adam Sobel, an earth and environmental sciences professorat Columbia University. He said that Emanuel'swork may be the first to deploy an approachthat comprehensively analyzes the changing probability ofa storm hitting over time.

Related: Determining business interruption losses afterHurricane Harvey

The report's conclusion is "stunning" and consistentwith Houston's recent spate of floods, but raises more questionsthan it answers, Sobel said. For instance, Emanuel wasunable to determine why Hurricane Harvey stalled overHouston.

There are no facts about the future, the saying goes. Untilthere are, physics models tuned with historical data arethe next best thing. 

Related: How insurers and reinsurers track live hurricanesin real-time