by David Keys
Indeed, the most remote of them have solar orbits that take them 750 times farther out into space than Pluto, the most distant of the sun’s planets. Out of the ten thousand billion comets estimated to be orbiting the sun, only a few thousand come within even three hundred million miles of the earth—or the sun!
Very, very few ever actually hit our planet. It’s estimated that a comet of the size required to generate mid-sixth-century-style climatic chaos collides with the earth on average only once every five hundred million years. But although comet impacts may be extraordinarily rare, they still have to be considered as theoretically possible culprits for the 535 catastrophe.
Comets are 70 percent frozen water, 15 percent frozen carbon monoxide and other gases, and 15 percent dust, stones, and possibly even boulders. Most comets are simply frozen lumps of ice and dust with temperatures as low as minus 454 degrees Fahrenheit. But a tiny number of them come briefly close enough to the sun to begin to “melt” (technically, they sublimate). Three hundred million miles out in space they then begin to form atmospheres—derived from their “melting” frozen-gas bodies. By the time the comet is 250 million miles from the sun, elements of that atmosphere and the freed-up dust within it begin to be pushed out to produce one (or sometimes two) “tails,” which can be up to 100 million miles long. It is the physical pressure of light (photons) on tiny dust particles that, quite literally, pushes the dust outward to form a tail. The gas molecules making up the atmosphere form the rest of the tail system by being given an electric charge and then by being carried along by ionized atomic particles shot out by the sun (the so-called solar wind).
The only comet impact ever actually witnessed and recorded by scientists was the collision of the comet Shoemaker-Levy 9 with Jupiter in 1994. In that event, a 2.5-mile-diameter comet nucleus broke up while temporarily orbiting the giant planet, and the twenty-one resultant fragments plunged at 2,200 miles per hour into Jupiter’s atmosphere. There was then a huge explosion that created a nuclear-style mushroom cloud extending two thousand miles above the cloud tops, billowing spectacularly into outer space, way outside the Jovian atmosphere. This massive cloud was caused by simply the largest fragment—a lump of ice and dust just half a mile across.
If it was a four-mile-diameter comet that caused the A.D. 535 event, the explosion would have been twenty times as great as the one on Jupiter in terms of energy release!
Although the comet or asteroid scenarios would explain the size and nature of the sixth-century catastrophe, there are a number of serious objections to either of these explanations.
First of all, an asteroid (or comet) impact, by definition, releases virtually all its energy in less than a second. The explosion caused by a large asteroid or comet hitting the ocean at high speed would have caused a vast circular wave, consisting of hundreds of cubic miles of water, to rear up around the impact site, penetrating deep into the stratosphere. The impact itself and the collapsing wall of water would then have sent a series of huge tidal waves hurtling across the ocean. Ninety-nine percent of each wave structure would have been deep below the ocean surface—stretching up to 3 miles down. Each visible wave would have been merely the surface symptom of a massive movement of subsurface water. Thus, at any one time, the wave motion would have involved thousands of cubic miles of water. On the surface, well away from the impact site, say two thousand miles, the tidal wave would probably have been only fifty feet high.
But as the wave approached land and entered shallower water, the percentage of the wave structure above sea level would have massively increased. Indeed, by the time the largest tidal wave reached the thousands of miles of coastline surrounding the ocean, it would have been perhaps three hundred to nine hundred feet high, enough to devastate hundreds of thousands of square miles of coastal land. Only where high cliffs or coastal mountains blocked the tidal wave’s path would the devastation have been limited. But where coastal plains were low-lying or where deep river valleys stretched inland through coastal mountains, the destruction would have been total, with the tidal wave penetrating dozens or even hundreds of miles inland in some areas.
The problem is that it would be nearly impossible for such a coastal catastrophe to have gone unnoticed by modern archaeologists, geologists, and historians. A tidal wave of such proportions would have rivaled Noah’s flood in any legend, would have been recorded in horrified terms by any literate societies affected, and would have been detected by archaeologists and geologists on any archaeological and geological site anywhere along thousands of miles of the relevant ocean’s coastline.
It is virtually unthinkable that an impact event of that magnitude only fifteen hundred years ago could be unknown to science today.
OPTION THREE: A MASSIVE VOLCANIC ERUPTION
Another clinching piece of evidence that points away from a cosmic impact explanation and toward the third option—a volcanic one—is this: Buried up to sixteen hundred feet below the surface of the Greenland and Antarctic ice caps is a telltale layer of ice contaminated by sulfuric acid of volcanic origin that was almost certainly associated with the twelve- to eighteen-month-long sun-dimming event of 535–536 and the subsequent climatic chaos.
Back in 1978, a joint Danish-Swiss-U.S. scientific team landed on the south Greenland ice cap in several large freight aircraft specially fitted with giant skis. The planes carried massive quantities of equipment, including generators, refrigeration units, prefabricated living quarters, and a huge drill.
This latter piece of hardware was used to extract more than a mile of ice core in about six-foot lengths. Working in temperatures as low as minus 22 degrees Fahrenheit, engineers and scientists drilled in three shifts, twenty-four hours a day, going deeper and deeper into the ice cap at roughly 400 feet per week.
Then, early in the second year of the operation, after just a few weeks of drilling, the team extracted some lengths of core covering the second quarter of the sixth century A.D. Back in a laboratory at Copenhagen University, chemical analysis of this sample revealed that there had been two substantial volcanic eruptions. These same eruptions were then detected in a second core drilled in summer 1990 in central Greenland.
Because the dating of Greenland ice cores at that time depth is only roughly accurate (say, to within five or eight years, depending on the core concerned), the two cores each gave slightly different dates for the same sulfuric acid layer. Dates are determined by simply counting back annual layers of snow—so unusually high precipitation can sometimes appear to add extra years, making an acid layer seem marginally older than it is.
The annual layers of snow (which under many tons of pressure turn into ice) are detected by measuring the normally regular annual variations in the percentage of snow consisting of so-called heavy water. Most water molecules, in liquid or frozen state, consist of two atoms of hydrogen and one oxygen atom with an atomic weight of 16. However, around 0.2 percent of water consists of molecules made up of two hydrogen atoms plus one oxygen atom of atomic weight 18. But the actual percentage of heavy water being precipitated onto a given point on the earth’s surface at any given time depends upon the weather. The proportion goes down to around 1,960 parts per million in very cold conditions and up to 2,000 parts per million in very warm conditions. In a polar environment the amount ranges from 1,960 parts per million in winter up to 1,975 parts per million in summer. Thus, by studying the regular (normally annual) rise and fall of heavy-water content in ice cores, scientists can construct an ice-core chronology and obtain dates for volcanically derived acid concentrations detected in the core. The system is accurate to within a few years, but unseasonable temperature fluctuations or substantially reduced levels of precipitation can distort the dating, especially at relatively large time depths.
For eruption one, the high-altitude GRIP core gave an apparent date of 527, while the lower-altitude DYE 3 core (three hundred miles to the south) yielded an apparent date of 530. The volcanic explosion must have been very substantial, as evidence fro
m the GRIP core shows that acid-rich snow was falling at the GRIP site in Greenland for more than two years and at the DYE 3 site for at least a year.
For eruption two, the high-altitude GRIP core provided an apparent date of 532, with acid snow falling on the site for just over a year. For this same eruption, the DYE 3 core yielded an apparent date of 534 and evidence of acid snowfalls of around four months.
The final evidence, however, comes from ice cores drilled ten thousand miles to the south, from deep inside the Antarctic ice cap. There, 660 feet below the windswept surface, scientists discovered evidence of a truly massive volcanic eruption.5 The ice-core material revealed that acid snow had cascaded down on the Antarctic for at least four years running. From the Antarctic ice cores at that time depth there are no accurate dates available—only rough, fifty-year-long ranges of dates. All that can be said is that the four-year-long acid snow episode recorded in the core occurred sometime between 490 and 540.
But by examining the acid traces left by other first-millennium A.D. eruptions it is possible, through a process of elimination, to conclude that the four-year acid episode must have been associated with the climatic catastrophe and probably with the 535 eruption. This is because the two chronologically adjacent Antarctic acid episodes were, respectively, in the fifty-year brackets 231–281 and 614–664—and because the four-year event that occurred in the 490–540 bracket is by far the biggest event recorded in Antarctica for the whole of the first millennium A.D.
It is very likely, therefore, that the Greenland and Antarctic ice cores signal the same atmosphere-polluting climate-changing event recorded historically and in tree-ring terms for 535–536.
Alternatively, though much less likely, the four-year Antarctic event could record a second, totally separate (or, indeed, connected) volcanic eruption destabilizing Southern Hemisphere climate in around 540—and helping to further destabilize Northern Hemisphere weather, already thrown into chaos by the big 535 event. The second Greenland acid signal (532–534 +/− 5–8 years) could conceivably, in this alternative though less likely scenario, have been generated by such a 540 second eruption.
But where did the 535 eruption take place? Which volcano was the culprit?6
31
T H E B I G B A N G
The first clue as to the location of the 535 eruption is the fact that the event is probably recorded in both the Greenland and the Antarctic ice cores. This double record indicates that the eruption must have been within the tropics; otherwise it would not have shown up as an acid spike at opposite ends of the world. It shows that acid snow was falling on both ice caps and had to have been delivered there by the two totally separate high-altitude wind systems that operate in the Northern and Southern Hemispheres, respectively. Only a tropical eruption could have achieved this to any substantial extent. However, the fact that acid-snow deposition took place for at least twice as long in Antarctica as in Greenland suggests that the eruption occurred in the southern rather than the northern tropics.
Luckily, there are only a limited number of active volcanic areas in the southern tropics: East Africa (including the Comoros Islands), the central Andes, the Galápagos Islands, and the huge chain of volcanoes stretching five thousand miles from the tiny Pacific island of Samoa to the large Southeast Asian island of Sumatra.
Judging by the massive climatic effects and the longevity of the acid spike in Antarctica, the eruption must have been absolutely enormous—bigger, probably substantially bigger, than the 1815 Tambora eruption on the Indonesian island of Sumbawa, which created a caldera over three and a half miles in diameter.
In East Africa, South America, and the Samoa-Sumatra chain, there are fewer than twenty known calderas that are big enough to be candidates—and fourteen of these are in the Samoa-Sumatra complex. However, the eruption dates of five of these are known not to have been in the sixth century.
The search can be further narrowed down by carefully examining the chronology and details of the climatic effects. Of those areas where records were made, the Far East was hit first and worst. That information—together with the fact that 70 percent of the candidate volcanoes lie in the Samoa-Sumatra complex—strongly suggests that the culprit erupted somewhere in that chain. By pure good luck, the suspects can be narrowed down still further, for buried deep in the Chinese History of the Southern Dynasties is a reference to what appears to have been a vast explosion in February 535.
The actual text says that “there twice was the sound of thunder.” Nothing very extraordinary about that, you might think. However, the entry becomes potentially significant because it is one of only three references to “thunder” in the whole of the first half of the sixth century—and is the only reference that does not describe the “thunder” as part of a massive storm or as being associated with lightning. In any fifty-year period, the Chinese chroniclers would have been able to record thousands of thunderclaps—but they didn’t. Only the very severest thunderstorms or those thunderclaps that were unexplainable or mysterious would have been recorded.
It is also the only one of the three thunder incidents in which the chroniclers specifically noted that it was a double event—that the sound was heard twice. Again, this is potentially significant.
Of course, the mere fact that the explosion appears to have been heard in China does not, on its own, pinpoint any particular volcano. But, fortunately, the brief Chinese account also reported that the two bangs came from the southwest. The chronicle was written by scribes based in the south Chinese imperial capital, Nanjing. So if the Chinese account is to be believed, and if the two bangs were indeed volcanically generated, the culprit volcano must have erupted somewhere southwest of Nanjing.
At first sight, that appears to present a problem, because the nearest suitably sized calderas in that direction are located twenty-eight hundred miles away, in the Sumatra/western Java area—presumably too far away to be within earshot. However, volcanic explosions can indeed be heard for thousands of miles. In 1883 Krakatoa was heard four thousand miles away. In the 1815 Tambora volcanic explosion, the sound traveled at least two thousand miles.
The sound of a volcanic explosion is actually transmitted by being bent and bounced through the atmosphere up to twenty times. Those parts of the sound-wave front that travel straight up from the volcano disappear into the far outer atmosphere and are lost. But most parts of the wave move out from the explosion at angles less than the vertical. As wind speed increases with height and because, starting at around seven miles altitude, air temperature also increases, sound waves are bent in much the same way that light is when it passes through a prism.
The bending process is often so pronounced that after some 150 to 200 miles the waves have been refracted to such an extent that they hit the earth’s surface, either the land or the ocean. The low-frequency sound then simply bounces off the surface like a huge echo and heads back into the atmosphere, which bends it a second time. The process is then repeated again and again until finally the energy of the sound wave is dissipated into the ocean, the ground, or the air.
For the sound to travel the twenty-eight hundred miles from a volcano in the Sumatra/western Java area to Nanjing would have taken around four hours, and not all the audible sound would have been lost. In fact, there are several factors that could have substantially strengthened it.
First of all, large volcanic eruptions produce unusually high percentages of low-frequency (i.e., long-wavelength) noise, which is absorbed less well by the atmosphere than is high-frequency (short-wavelength) noise.
And second, because of the refractive effect of the atmosphere, the different parts of the wave front would have traveled along different paths and been accelerated by different wind speeds. The short though very loud sound of a volcanic explosion would therefore have been lengthened into a sound lasting several minutes. However, because so many atmospherically bent sound waves would be echoing off so many parts of the earth’s surface, some waves would reconverge at various po
ints, thus reinforcing each other to produce a plural number of sound peaks within a less audible lengthened-sound phenomenon.
So, if the Chinese chroniclers’ mystery double bang emanated from a volcanic eruption thousands of miles to the southwest, which volcano could it have been?
* * *
Bearing in mind that the eruption had to have occurred in the southern tropics, the area pinpointed by the Chinese account narrows the field down to the southern Sumatra/western Java part of the Samoa-Sumatra volcanic chain.
Significantly, there is only one known caldera of appropriate size and vintage in that relatively small (six-hundred-mile-long) area. It surrounds the site of no less notorious a volcano than Krakatoa, the island mountain that brought death and destruction to Java and Sumatra in the 1880s. Could an earlier, bigger eruption of Krakatoa have been responsible for the catastrophe that tormented the world in the mid–sixth century A.D. and changed its history forever?
