During the past several thousand years, the earth's climate has certainly not changed radically. But we can't take that for granted. It's entirely possible civilization arose because of an anomalous period of climate stability. Should the climate change significantly, civilization would be threatened.
You may think I'm referring to human-caused global warming. I'm not. Although the extremely rapid (in terms of geologic time) increases in carbon dioxide levels may lead to disaster, I'm referring to natural processes that are outside of our control. I've already covered one – the meteor threat (although we can control that to some extent). Here's another:
One of the most interesting and important theories relating to long-term climate changes was developed by the Serbian astronomer Milutin Milankovitch (1879-1958). In an attempt to understand the Ice Ages, he studied long-term variations in the Earth's orbit. Milankovitch determined that there were at least three variations with significant climatic effects:
"Change in eccentricity" means the orbit becomes more elliptical – the Earth-Sun distance changes as the Earth moves in its orbit. When eccentricity is low, the orbit is nearly circular; when it is high, the orbit is more elliptical. The closer the Earth is to the Sun, the more heat it receives from the Sun. As the earth moves farther from the Sun, it receives less heat. Currently, there is a difference of only about three percent (five million kilometers or three million miles) between the closest distance (astronomers call it perihelion), and the farthest distance (called aphelion). Recall that the average Earth-Sun distance is roughly 93 million miles, or 150 million km.
Each second, the Sun generates heat – an enormous amount of it. The heat moves away from the Sun, at the speed of light. As the rays of heat move outward, think of them as defining a sphere. As the sphere expands, the Sun's heat is spread over a larger and larger area. So any given object (such as the Earth) will receive a smaller fraction of the total. The area of a sphere of radius R is given by
where π is approximately 3.14159.
We don't actually have to calculate this area. All we need to know is that the area increases as the square of the radius. So if we increase the distance from the Sun by three percent, the area – over which the Sun's heat is spread – goes up by six percent (in other words, the square of 1.03 is approximately 1.06). Since the Sun's heat is spread out over six percent more area, the Earth (which doesn't change in size) will then get six percent less of that heat. So, during each year, when the Earth is farthest from the Sun during its orbit, it gets six percent less heat than when it's closest to the Sun.
Is this why we have seasons? No. As it turns out, the Earth is closest to the Sun around Jan 3, near the onset of winter in the Northern Hemisphere. It is farthest from the Sun about July 4, near the onset of summer (again, in the Northern Hemisphere). So the Earth is, because of being closer to the Sun, actually warmer in winter (in the Northern Hemisphere) than it would otherwise be.
The reason for the Earth's seasons is the Earth's tilt on its axis. When the Northern Hemisphere is tilted toward the Sun, as it is during the summer, it's warmer. It gets colder during the winter, when it's tilted away from the Sun. During a cycle that lasts about 40,000 years, the earth's tilt on its axis varies between roughly 22 and 25 degrees. As the tilt increases, the difference between seasons becomes greater. More tilt means warmer summers and colder winters. It's believed that cooler summers (less tilt) allow snow and ice to last from year to year in high latitudes, which allows the buildup of ice sheets. As ice sheets build, they provide positive feedback in the climate system. (This means they increase the tendency to build glaciers – snow and ice reflect the sun; that means more snow and ice forces more sun to be reflected; that causes cooler weather; that causes more snow and ice, and so on . . . )
Likewise, if there were already large glaciers on the Earth, and the tilt started to increase, causing warmer summers in the Northern Hemisphere, this would tend to melt the glaciers. This would cause positive feedback for warming – less snow and ice would cause less sunlight to be reflected; this would cause warmer conditions; this would cause fewer glaciers, less reflection of sunlight, warmer conditions, fewer glaciers . . . The Earth is currently in the middle of one of these cycles of tilt.
The final effect is axial precession. Currently the Earth is closest to the Sun (in perihelion) near the start of winter in the Northern Hemisphere; it's farthest from the Sun near the start of summer. This decreases the severity of the seasons in the Northern Hemisphere, and increases them in the Southern. Over about 20,000 years, the date of perihelion goes in a complete cycle. So about 10,000 years ago, the distance of closest approach occurred during the start of summer (in the Northern Hemisphere).
Now let's put these three effects together. First, I have referred to what happens in the Northern Hemisphere. Why not the Southern? Because the Southern Hemisphere is mainly water, its seasons are milder than those of the North. Water has a high heat capacity (it takes a long time to heat it or cool it), so it spreads out seasonal extremes. It's entirely possible that the Earth wouldn't have Ice Ages if land and ocean were equally distributed in both Hemispheres. Also, smaller continents, surrounded by more water, would lead to less seasonal extremes. Large continents would heat or cool much more, leading to more extremes. So, many millions of years ago, when the continents were distributed differently (and were of different sizes than today) orbital variations would have affected the climate differently. It's also possible that the Earth's orbit has additional variations, on times scale much greater than those Milankovitch found.
If we believe that extreme seasons (in the Northern Hemisphere) prevent glaciers from forming, and moderate seasons allow them to form, here's what happens. The most severe seasonal extremes would occur under three conditions. First, the Earth would have to be in a very elliptical period, so that the Solar radiation received varied by 20% to 30% in the course of a year. Second, the aphelion (farthest distance from the Sun) would have to occur during winter. This would make Northern Hemisphere summers very hot, and winters very cold. Thirdly, the tilt of the Earth's axis would have to be a maximum, creating hotter summers and cooler winters.
The most moderate seasons would occur when the orbit is the least elliptical (as it is now), the aphelion occurred during summer (as it does now), and the tilt is the least (not true now; we're in the middle of the tilt cycle). So we're currently much closer to ideal conditions for starting another Ice Age – assuming moderate seasons cause Ice Ages – than to conditions which would lead to the end of one! Considering the last Ice Age ended only about 10,000 years ago, maybe that's not surprising.
Now I've stated what extreme seasons would do. But this isn't hypothetical. Not only will these seasonal extremes occur in the future, they already have. Is it possible that civilization exists now only because of some happy (orbital) accident?
Milankovitch's theory shows us several important things. If the earth's orbit were very elliptical, winters might be too cold (and/or summers too hot) to sustain life. Extreme summer heating could lead to exceptionally destructive hurricanes – or something even worse. If Earth were to have a large tilt, this would also lead to severe seasonal extremes, with possible adverse effects on civilization or the development of intelligent life. Extreme variations in tilt could also be bad. The proportion of land to water, the distribution of continents on the earth, and the sizes of the continents – all have effects on climate. It's not clear that all planetary systems – even those that have planets in a "habitable zone" – would have planets that meet the requirements for complex life, or have a stable climate that could lead to the evolution of intelligent life.
Also, as noted above, currently Earth's orbit is such that we are close to the most moderate climate. In about 50,000 years, the Earth's orbit will be at its most elliptical. The change in heating by the Sun will vary, over the course of a year, by 20 to 30 percent. Also, the distance of closest approach to the Sun will then occur during the start of summer in the Northern Hemisphere. So the variation between summer and winter will be much more extreme than it is today. This could have severe impacts on civilization. Perhaps it already has – 50,000 years ago the Earth had this highly elliptical orbit. Perhaps conditions were too extreme for civilization to happen?
Sadly, while Milankovitch developed his theory in the 1920's, he didn't live to see it accepted (he died in 1958). It wasn't until 1976 that an study in Science found that Milankovitch's orbital variations corresponded to periods of glacial creation and destruction. The study's authors used deep-sea sediment cores to determine temperature changes going back 450,000 years. They found that major climate changes were indeed closely associated with the orbital variations that Milankovitch discovered. The National Academy of Sciences embraced his model in 1982.
Some other Milankovitch references/articles:
Britannica
Live Science
Tesla Society