Dancing With Myself
Page 32
These are North American evidence of what I like to call “close encounters of the fourth kind.” The Earth must have seen many events like them in its long history. Only the weathering effects of the atmosphere, plate tectonics, and biological organisms save Earth from being as heavily cratered as the Moon.
Small meteors hit the Earth every day, and are burned up in their passage through the atmosphere. How often do meteorite impacts of a substantial size occur? This is an area where there are very few solid data to guide us, but we can make plausible estimates by putting together several apparently unrelated facts. First, the Earth and Moon are close neighbors in celestial terms, and they should encounter about the same number of meteorites, once we make allowance for the Moon’s smaller size. We can count the Moon’s craters, and their number suggests that an object big enough to make a crater a mile across will hit every hundred thousand years or so. A comet fragment twenty meters in radius would do it. Allowing for Earth’s larger size, something that big should hit us about every ten thousand years.
Statistical analysis of bodies in the asteroid belt also provides a rule-of-thumb, saying that for any asteroid of a particular radius, there will be ten times as many with one-third that radius. We are going to assume that the same distribution law applies to comets, too. (We have to—data on comet nucleus sizes are too sparse to establish any frequency/nucleus-size relationship.) The rule-of-thumb can readily be converted to a general formula that tells the number, n, of bodies of any radius, r, thus:
n = N10-log(r/R)/log3 (4)
where N and R are any pair of known values.
Assuming the impact of an object twenty meters in radius every ten thousand years, the size/number relationship of equation (4) allows us to calculate the frequency of an impact of any size of body, and we already know how to compute the associated energy release. Table 2 shows the average time between impacts for different sizes of cometary bodies.
The table tells us that there should have been only four impacts of something the size of Halley’s Comet since the Earth was formed. Given the huge energy release this implies, that’s just as well. Conceivably, four such events correspond to major species extinctions in Earth’s history. (A detached attitude to such calamities is hard to achieve, but possible. I was driving James Lovelock, originator of the “Gaia” concept [more about Gaia later], down to the Museum of Natural History in Washington. On the way we somehow got onto the subject of all-out nuclear war. Lovelock surprised me very much by remarking that it would have very little effect. I said, “But it could kill off every human!” He replied, “Well, yes, it might do that; but I was thinking of effects on the general biosphere.”)
Even if we keep our missiles in their silos and submarines, the planet will have seen a “nuclear war” energy release from a comet or meteorite impact an average of every two million years; a one-megaton hydrogen bomb equivalent every two thousand years; and a Hiroshima-sized event every 130 years. Historical evidence suggests that these rates are on the high side—not surprising, considering the tenuous nature of some of our assumptions. On the other hand, the reluctance of our ancestors to accept the idea of meteorites suggests that any fireball occurring, say, four hundred years ago, might have been misinterpreted6—or, in much of the world, not recorded.
It would be nice to think that explosions mimicking an atomic bomb in violence are rather rare. The response of a nervous nation to a Hiroshima-style fireball over one of its major cities is hard to predict.
Table 2: Size, Frequency, and Effects of Comet Impacts
Size of body
(radius in meters)
Frequency of
occurrence (yrs)
Energy release
(ergs)
Energy release
(Megaton H-bombs)
2
.5
128
8.8 × 1020
0
.02
5
550
7.1 × 1021
0
.17
10
2,340
5.7 × 1022
1
.4
20
10,000
4.5 × 1023
11
40
43,000
3.6 × 1024
86
Nuclear wars
(1 nuclear war =
25,000 megatons)
60
100,000
1.2 × 1025
286
100
292,000
5.7 × 1025
1,350
150
680,000
1.9 × 1026
4,520
250
2,000,000
8.8 × 1026
21,000
0
.84
500
8,500,000
7.1 × 1027
169,000
6.
.8
1,000
36,000,000
5.7 × 1028
1,360,000
54
.3
2,500
250,000,000
8.8 × 1029
21,000,000
838
5,000
1,060,000,000
7.1 × 1030
169,000,000
6,760
10,000
4,500,000,000
5.6 × 1031
1,347,000,000
54,000
The Hiroshima atomic bomb was about 20 kilotons TNT equivalent. The 2.5 meter comet fragment releases as much energy as one Hiroshima bomb, and the 5-meter fragment as much as nine such bombs. A 2.5-meter fragment impact can be expected every 128 years. Objects of this size and smaller will produce a fireball as they burn8789 up in the atmosphere, but normally will not reach the surface of the Earth.
8.STELLAR EVENTS
The energies of a nuclear war or a cometary impact are huge on the everyday scale of Earthly events, but they are minute compared with the power production of even the smallest and dimmest stars.
The Sun, a rather average G2-type dwarf star, emits 3.9 × 1033 ergs of radiative energy per second—that’s four million nuclear wars a second. Fortunately the Earth intercepts only a tiny fraction of the solar bounty, roughly one two-billionth.
The question now is, can this energy change enough
to threaten the survival of life on Earth?
The obvious danger in this case might seem to be an excess of radiation—of frying, rather than freezing—since there are no signs that the Sun is likely to go out for many billions of years. But could the Sun become much brighter? The sunlight delivered to the Earth is known as the “solar constant,” and it is about 0.14 watts per square centimeter. Can the solar constant change, because of, say, a very large solar flare?
Well, over very long periods the solar constant has certainly changed. Life has existed on Earth for about three and a half billion years; and in that time, the solar constant has increased by at least thirty percent. If Earth’s temperature simply responded directly to the Sun’s output, two billion years ago the whole Earth would have been frozen over.
But in fact, the response of Earth’s biosphere to temperature changes is complex, apparently adapting to minimize the effects of change. This is part of the whole Gaia concept, of life on Earth as a giant mechanism that regulates its own environment in an optimum manner. For example, as temperatures go up, the rate of transpiration of plants increases, so the amount of atmospheric water vapor goes up. That means more clouds—and clouds reflect sunlight, and shield the surface. In addition, increased amounts of vegetation reduce the amount of carbon dioxide in the air, and that in turn reduces the greenhouse effect by which solar radiation is trapped within the atmosphere. There are many other processes, involving other atmospheric gases, and the net effect is to hold the status quo for the benefit of living organisms.
However, there is a big difference between a 30% change that takes place over three and a half billion years, and one that takes place overnight. We need to know if a rapid change is possible.
The picture is a little bit confusing. The Sun emits almost all of its light energy at ultraviolet, visible, and infrared wavelengths (99% of the total between 0.276 and 4.96 micrometers). Measurement of the solar constant over this range shows very little change. On the other hand, there is a definite cyclic variation in solar output at X-ray and radio wavelengths, corresponding to the eleven-year cycle of sunspot activity and to other, longer periods. The fraction of energy emitted at these wavelengths is small, but the effects are certainly not negligible.
For example, although the Ice Ages took place before recorded history, there were two well-documented “Little Ice Ages,” one from 1460 to 1550, the other from 1645 to 1715. These periods, known as the Spörer Minimum and the Maunder Minimum respectively, occurred at times when there were almost no sunspots on the Sun. Flamsteed, the first British Astronomer Royal, hardly saw a sunspot in forty years of observations. Isaac Newton, whose lifetime (1642-1727) neatly overlaps the Maunder Minimum, was in a similar position.
Conversely, the so-called “Grand Maximum” from 1100 to 1250, when Greenland was settled, has been studied by J.R. Eddy using carbon-14 dating of tree rings. It proves to be a period of prolonged sunspot activity and a warm Earth.
There is thus no doubt that quite small changes in solar output can have significant effects on the Earth’s climate. The natural (but wrong) conclusion is that the major Ice Ages were caused by correspondingly larger changes to the solar constant. Actually, Milankovitch has produced convincing evidence that the Ice Ages correspond to changes in the Earth’s orbit, rather than changes in solar output.
Theories of stellar evolution tell us that there have been slow, steady increases in solar heat production, over billions of years. But history offers no evidence of large, sudden excursions of the solar constant from its usual value. Sol is a remarkably stable furnace, having less effect on Earth’s climate than the eccentricity of the Earth’s orbit around the Sun.
As a competitor with meteors and nuclear wars to effect an abrupt end to human affairs, solar energy variation seems to be a non-starter.
9.SUPERNOVAS
With supernovas, we move into the big league. If Sol were to turn into a supernova, its light production could increase by a factor of a hundred billion, to 4 × 1044 ergs a second.
This sounds like a lot—it is a lot—but it is only a tiny part of the supernova’s total energy production. Between ninety and ninety-nine percent of the energy in the explosion is carried off by neutrinos. Of the remainder, ninety to ninety-nine percent is in an exploding shell of matter, blasted outward at a twentieth of the speed of light. Only between one percent and one one-hundredth of a percent of the energy is emitted in the form of radiation.
The neutrinos appear at the moment of the explosion, whereas the emitted light and high-energy particles increase and then decrease in intensity over a period of weeks or months. The 1987A supernova in the Large Magellanic Cloud produced and emitted an estimated 1058 neutrinos in just a few seconds, and they carried off with them 3 × 1053 ergs—equivalent to one-tenth of the mass of the Sun.
Neutrinos interact with normal matter hardly at all, which is why they readily escape from the center of the supernova. A neutrino can pass through several light-years of lead before being captured. However, the number emitted in a supernova explosion is so large that the neutrinos alone would kill a human a billion kilometers away. If Sol were to became a supernova, we would be wiped out by the neutrinos shortly before we were vaporized by the flux of radiation.
Before we worry too much about that event, we ought to note that according to today’s theories Sol cannot become a supernova. There are two types of supernova. Type I occurs only in multiple star systems, when a massive white dwarf receives enough matter from a stellar companion to render it unstable. Type II occurs when the core of a giant star, ten or more times the mass of the sun, collapses. (More details of what happens in both Type I and Type II supernovas are given in the article, “Something for Nothing.”) Since our sun is neither a binary star nor a giant star, we seem to be safe. On the other hand, how much do you trust today’s theories?
In our local stellar neighborhood there are candidate multiple star systems and giant stars. It is possible that one of those could produce a supernova. Is it then conceivable that the event would be energetic enough to destroy life on Earth, or at least produce a huge perturbation comparable with a nuclear war or a large meteor impact?
Let us look at the numbers.
The closest multiple star system to Earth is also the nearest star system, Alpha Centauri. It is a very unlikely candidate to become a supernova, but if it were to do so, would it harm the Earth? Alpha Centauri is 4.3 light-years away, more than 270,000 times as far as the Sun, and distance is the best protection.
The best protection in this case is not quite enough. If one of the three stars in the Alpha Centauri system became a supernova and produced 4 × 1044 ergs per second in the form of light, it would shine a third brighter than the Sun for a few weeks. The increased heat alone might not kill us, but the sleet of high-energy particles, carrying ten times as much energy as the light, would be even more destructive. However, we would have plenty of warning of the coming particle storm, since the radiation from Alpha Centauri would precede the particles by three-quarters of a century. Digging would become the new international pastime.
I can’t help wondering how the world would react to the idea that, following a terrible time of heat and chaos, the worst was still to come. Would people believe the scientists’ statements? Would they be willing to begin preparation now, for an event so far in the future that most people would not be there to experience it? Or would they shrug and say, “Let them handle it when it happens—it’s their problem”? We have seen a lot of that attitude towards environmental pollution.
Supernovas vary in the violence of their explosions, but anything closer than 50 light-years might produce severe effects on the Earth, as much as if we had a sudden increase of ten percent in the solar constant.
Again, the probability of such an event is more important than the possibility. To determine this, we have to know the rate of occurrence of supernovas. The easiest place to lo
ok for supernovas is not in our own Galaxy, since much of that is obscured from us by interstellar dust clouds. It is better to look at neighboring galaxies, such as Andromeda, and count supernovas there. That exercise suggests that a supernova occurs maybe every century in a galaxy the size of ours, with an uncertainty on that number of at least a factor of two. Since the Galaxy contains about a hundred billion stars, and since there are about 1,000 star systems within 50 light-years of us, we can expect a supernova within this distance only once every 10 billion years. This is a simplistic argument, neglecting the different types of stellar populations, and where they lie within the Galaxy, but again we are looking for ballpark figures.
The closest and brightest supernova in recorded history occurred in A.D. 1054, and the remnant of that explosion now forms the Crab Nebula. The Crab supernova was bright enough to be visible during the day, but it had no harmful effects on Earth. It lies about 6,000 light-years away from us. We can expect a supernova this close or closer every 6,000 years. It is a little disturbing that the event actually occurred less than a thousand years ago, and it suggests that either supernovas are more frequent than we think, or more likely we are in a galactic region that favors supernovas. If a supernova as close as the Crab nebula occurs every thousand years in our galactic neighborhood, then we can expect a supernova within 50 light-years every couple of billion years.
One final question is of interest: how often will a supernova deliver the energy equivalent of a nuclear war to the Earth, in radiation and particles? Taking the total radiation and particle energy of a supernova as 1050 ergs, it is easy to calculate how far away the exploding star can be if the Earth is to intercept the necessary 1027 ergs. The answer is a little over a hundred light-years. Making the same assumption as before about the frequency of supernovas in our galactic neighborhood, this will happen every two and a half billion years. Apparently supernovas are not a major danger to the human race. (But of course, statistics being statistics, a nearby supernova could explode tomorrow.)