Global Warming, Ice Ages, and Sea Level Changes:
What certain special interest groups don't want you to know
Last Modified July 10, 2010
There is a lot of concern today about global warming and controversy regarding the degree to which modern fossil fuel use is affecting the world's climate. This article is intended to provide some related geologic and astronomic information to which most of the general public has never been exposed. Hopefully, this will provide some background for making informed personal decisions about this controversial subject and alleviate fears that civilization is doing anything to Earth's climate that hasn't happened completely naturally in the past.
First let me say that global warming is real. So is global cooling. Scientists who examine the climate record see both upward and downward trends in the data, depending where in the geologic record they are looking. Every grade school student knows that the Earth has gone through countless cycles of Ice Ages and interglacial periods, and paleoclimatologists have found that these shifts in global climate have happened on a very predictable schedule. Currently, the Earth's average temperature is rising, and the effects of this warming are noticeable within a person's lifetime. For example, areas in southern England can again grow vineyards, which they have not been able to do since before the 1700's (they were wiped out during the Little Ice Age, from which the Earth is still slowly recovering. see http://www.sciencenews.org/view/feature/id/5062/title/Global_Vineyard). Detailed US climate records from the past 30 years have shown that temperature increases are occurring (mostly in winter and nighttime). Modern tide gauges have shown that the Earth's average sea level is rising at about 23 centimeters per century, equivalent to the speed of rise about 5,000 years ago. There is no question that the world's temperature has been gradually increasing for thousands of years, right up to the present day. What appears different today is the rate of change, not the magnitude of the change. Sensationalized media reports never tell the public that even the worst-case climate modeling scenarios fall entirely within the temperature ranges that occurred naturally in the remote past before human civilization even existed.
Atmospheric and earth scientists have developed a theory that the generation of greenhouse gases, primarily carbon dioxide created by human consumption of fossil fuels, may be contributing to this temperature rise. Much research is currently being done to test this theory and the conclusion (so far) is that the observed rates of temperature change are consistent with the observed rise in greenhouse gas concentrations. Paleoclimatologists know from study of the past climate records that the climate's behavior is quite erratic, and the conditions we see today have been quite common throughout geologic history. Basically, the problem is that the predicted effect of human civilization on the world's climate could be less than the natural noise in the system. This makes definitive tests of the current theory difficult.
There is no doubt among qualified scientists that carbon dioxide in the atmosphere contributes to the greenhouse effect (where a larger than "normal" amount of solar energy is trapped as heat in the atmosphere rather than being radiated back into space). There is also no question that the burning of fossil fuel has added carbon dioxide to the atmosphere. Modern computerized global circulation climate models suggest that the amount of observed warming is approximately the same as the amount that theory predicts should be caused by the increased concentrations of greenhouse gases in the atmosphere.
The Scientific Method is the process by which the predictions of any theory, like Anthropomorphic Global Warming, are tested against new observations. When the predictions match the new observations, scientists continue to apply the theory and probe for inconsistencies. When new data is found that contradicts the theory's predictions, scientists are obligated to revise the theory to explain these new and unexpected observations. This revised theory then becomes the improved method for producing new predictions, which continue to be compared against new observations. It is by this method of theory revision that science advances. Sometimes, there are rival theories for the same phenomenon, in which case scientists gauge the ability of the various theories to predict observations as the basis of choosing the better theory. In plain language, even the most revered and accepted theories are subject to revision, or replacement, if their predictions fail to match observations.
How much Greenhouse Gas is in the Atmosphere? The following graph shows the proportions of atmospheric gases as reported by NASA. The vast majority of the atmosphere is Nitrogen and Oxygen, which account for about 99% of the atmosphere. Of the remaining 1% (the expanded pie slice on the right), almost all of it is Argon (0.93%). The tiny red slice is Carbon Dioxide, which accounts for 0.035% of the Earth's atmosphere. Recent research indicates that current Carbon Dioxide levels are about 25% larger than pre-industrial levels, and most of this excess is attributed to fossil fuel use. On days with normal or higher humidity, water is the third largest component gas in the atmosphere. Water vapor is a powerful greenhouse gas , but water vapor's concentration in the atmosphere is highly variable (as high as 3% at 86 degrees F, and no more than about 1/2% at 32 degrees F) and was therefore left out of the NASA data table. Of all the greenhouse gases, water vapor has the largest impact on global temperature because of its relatively high concentration in the atmosphere (at 100% relative humidity at 95 degrees F, water vapor is over 100 times more concentrated than CO2), far outstripping the effects of methane and Carbon Dioxide combined (water vapor accounts for 2/3 of today's greenhouse gas warming). In fact, without the warming effect of water vapor, Earth would have an average temperature below freezing (this actually occurred at the end of the Precambrian).
Carbon dioxide levels have been increasing since scientists started collecting the information.
This graph shows the measured CO2 concentrations since 1958 collected at the Mauna Loa Observatory in Hawaii. The saw-toothed cyclicity is due to the combined effects of seasonal variations in plant growth, decay, and fuel use in the northern hemisphere. Yearly low values occur near the end of Fall, when many plants stop growing or shed their leaves. CO2 increases in the winter and Spring as people burn wood and fossil fuels to keep warm and decaying plant material returns CO2 to the atmosphere. Each year, the CO2 increase in winter and spring is slightly more than the decrease in summer and fall.
The big questions facing climate researchers today are "What is 'normal' for Earth's temperature, and how large a deviation from 'normal' temperature behavior can human civilization cause"? To understand the effect human activities are having, first we must understand the 'background trend', in other words, what would be happening if there were no humans on Earth.
For generations, geologists have been trying to explain the obvious cyclicity of sedimentary deposits observed everywhere we look. The prevailing theories hold that this cyclicity primarily represents the response of depositional processes to the rise and fall of sea level. In the rock record, we see times when sea level was astoundingly low alternating with times when sea level was much higher than today, and these anomalies often appear worldwide. The image below shows the sedimentary sequences of the Grand Canyon and Rocky Mountain area. Note the strong cyclicity of the various rock types.
The above illustration by Ward Abbott is from the July 1998 AAPG Explorer. It shows the sedimentary units of the Rocky Mountains in the American west. Note particularly the alternation of the pink-colored mudstone/shale units with sandy units colored yellow and orange. The shaly units overlying sandstones record the rise of sea level. The colored columns on the left and right sides labeled "Sequence Order" show the interpretation of different sea level cycles by geologists. The 4th order cycles record changes on time scales of less than 2 million years.
The image below, scanned from an advertisement in the June 2005 AAPG Explorer, shows seismic (sound wave) reflection data from deep underground in an offshore area. The vertical axis represents several thousands of feet of sediment accumulation. The horizontal red lines are interpreted sedimentary cycles and the thickness of the lines indicates the relative periodicity of the cycles. The heaviest horizontal lines and the vertical bars bracket time intervals of tens of millions of years, and the finest horizontal lines represent cycles of 120,000 years or less. Even finer subdivisions representing 40,000 and 20,000 year cycles can be found on this seismic line. The hot pink line is a fault where the rocks have broken and slipped past one another (down on the left side). When the finest cycles get very closely spaced, they represent a time when sea level was very high and fluctuated very little, much like today.
Prehistoric Sea Level Changes During the depths of the last ice age 20,000 years ago, when hundreds of thousands of cubic miles of ice was stacked up on the continents as glaciers, sea level was 390 feet lower (~120 meters) than today. Locations that today support coral reefs were left high and dry, and coastlines were miles farther basinward from the present-day coastline. However, for the past 6000 years (long before mankind started keeping written records) the world's sea level has been gradually approaching the maximum level which we think the planet is capable of sustaining. Sea level could be higher than it is today. In fact, during the Pleistocene, between 118,000 and 133,000 years ago, sea level was for a short time (several centuries) about 19.5 feet higher than today, as evidenced by wave-cut notches along limestone cliffs in the Bahamas and New Guinea.
These terraces were left along the New Guinea coastline as sea level gradually fell at the beginning of the last Ice Age. During the previous interglacial period, sea level was clearly substantially higher than today, suggesting that there were very few glaciers left to melt. This implies that summer temperatures were also higher than today, preventing snow and ice buildup in areas prone to glacier development.
There are also Pleistocene coral reefs left stranded about 3 meters (~10 feet) above today's sea level along the southwestern coastline of West Caicos Island in the British West Indies . These once-submerged reefs and nearby paleo-beach deposits (see photo below) are clear and unambiguous evidence that sea level spent enough time at that higher level to allow the reefs to grow (this level is probably very close to present-day Earth's highest sustainable interglacial sea level).
This photo was taken at Boat Cove on the west coast of West Caicos Island in the small Caribbean nation of Turks and Caicos in 2004. The sedimentary features visible in this outcrop indicate that the rocks under the researcher's feet were deposited in the swash zone of a beach. This sediment is only about 120,000 years old and is currently at least 15 feet above present-day sea level. The location of this sediment is very strong evidence that sea level has fallen relative to the island since these beach sands were deposited.
The following graph shows a detailed compilation of sea level data points and interpretations for the last 11000 years for the western Atlantic Ocean and Caribbean.
Sea level interpretations for the past 15000 years. Today's date is on the right side of this graph. Data source for light blue line: Wanless & Davis, Carbonate Environments and Sequences of the Caicos Platform, 1989, American Geophysical Union, pg.4. The dark blue line is drawn across the top of dated samples that nature can create only in sites slightly below sea level (Acropora Palmata coral and mangrove swamp peat). Observations in south Florida show that sea level has risen and fallen 1-2 meters at least twice since 4000 years ago (Wanless & Davis, pg. 5), visible as a wiggle in the graph's spots between 5000 and 0 years. There is no reason to expect that the slow rise or the meter-scale oscillations of the past few thousands of years would not continue into the future.
The following two graphs show sea level variations in the Arabian Gulf since the end of the last ice age. The evidence the authors used are age-dated marine sediments now found at these heights above sea level.
In the figure above, the middle curve shows sea level changes since the Late Pleistocene. Sea level reached its lowest point (120 meters below today's sea level) about 16000 years ago. The upper line is the estimated temperature based on the temperature tolerance of plants that grew during those times. Note that temperatures were 5 degrees F warmer than today 6000 years ago. The lower line shows a generalized bathymetric profile of the Arabian Gulf bedrock illustrating the positions of wave-cut benches due to prolonged periods of sea level stillstand at those elevations.
In the figure above, notice that sea level was 3.5 meters above today's level between 5000 and 6000 years ago. The high amplitude sea level oscillations have been gradually damping out due to orbital forcing (a discussion of orbital forcing comes later in this paper)
The evidence for high ancient sea levels is the existence of beach and nearshore sediments far above today's sea level. The author took this picture of Late Pleistocene/Early Holocene subtidal nearshore sediments about 3 miles from the current-day Qatar shoreline. The top of these beds where the geologist is standing is about 2 meters above sea level today and would have been about a meter below sea level at the time of deposition.
To give a perspective of the longer-scale variation in sea level, the following graph shows the amount of ice volume that has been tied up in in glaciers for the past 140 thousand years. Please note that today's date is located on the right side of this graph (0 years ago).
This graph shows the amount of sea water that has been locked up in glacial ice over the past 140,000 years. Notice that sea level tends to go down in relatively small steps on a fairly regular cycle of 20,000 years, but rebounds to maximum in one big jump. The most recent rebound from an ice age started 20,000 years ago with sea level about 120 meters lower than today. The melting slowed down about 5,500 years ago, slowed again about 3,200 years ago, and sea level has been rising very slowly since that time. More detailed sea level measurements from the Florida coast show that sea level also oscillates at a comparatively "high" frequency of a few centuries, so the way sea level changes resembles the old saying "two steps forward and one step back". If the present-day interglacial period lasts as long as the Sangamon interglacial period 125,000 years ago, we should expect to see sea level starting to drop in another 5,000 years or so. This figure can be found athttp://www.grida.no/climate/ipcc_tar/wg1/fig11-4.htm.
The colored part of the following graph shows the changes in "eustatic" sea level (tectonic ground motion effects removed) over the past 30 million years. Note that the long-term trend shows an overall falling of sea level, but that the short-term trends are highly erratic. The point labeled 0M is today's sea level. The short-term curve over the most recent 1 million years is not shown here, but if it was it would show 10 excursions between the left side to the right side of the light blue area. Today's date is the top of the graph (Time=0).
Something is making sea level rise and fall on a global scale, and it has been going on for hundreds of millions of years.
What drives these sea level changes throughout geologic history? Temperature, of course! During times when the Earth is colder than present-day, water that evaporates from the sea gets trapped up on the continents as year-round glacier ice, and the sea level falls in equal measure. When the Earth's temperature rises, that continental ice melts and runs back to the sea and fills the ocean basins once again. We think that the total amount of water on Earth is relatively constant, so adding ice to continental glaciers always removes it from the ocean basins, and vice-versa.
How did the climate behave before the 1900s? The following graph shows the results of temperature estimates for the past 500 years using a variety of methods, including oxygen isotope analysis of lake sediments and coral growth rates. The spiky line shows modern thermometer measurements and clearly displays a frequency corresponding to the 11-year solar sunspot cycle (nine peaks since 1900). This picture represents a time duration which is only the tiniest fraction of the right end of the preceding ice volume chart. Notice that the temperature change over these 500 years is only 1 degree Centigrade (~1.8 degrees F), and only 1/2 degree in the last century.
Because this graph starts during the 1500s, it fails to illustrate that the Renaissance was an unusually cold period commonly called the "Little Ice Age" (~1400 AD to ~1850 AD). Prior to the Little Ice Age, temperatures were warmer than today. During the Little Ice Age there were famines, plagues, droughts, towns crushed by rapidly advancing glaciers, and other weather-related hardships. In the 1800s, when temperature finally began to increase enough to notice, relief from those devastating frigid conditions was heartily welcomed.
The spikes that occur once every decade in the instrumental records are the result of the 11-year sunspot cycle. Notice also the 60- to 80-year period cycle around the bold black running average line (peaks in 1860s and 1930s, and lows around the 1890's and 1970's).
This graph can be found at: http://www.grida.no/climate/ipcc_tar/wg1/fig2-19.htm
However, if we look further into the past, we can see that the temperatures we consider "normal" are actually anomalously low. Before 3.5 million years ago, the Earth's temperature never fell below what we consider normal and life on Earth thrived.
Please note that the little blue inset box has a temperature scale on the right side, and zero marks today's temperature. During the Eocene Optimum about 50 million years ago, temperatures were a whopping 15 degrees Celsius warmer than today. There were no Antarctic glaciers at that time. A temperature drop at the beginning of the Oligocene (~33 million years ago) allowed glaciers to form in Antarctica, but at about 24 million years ago the Earth heated up again and the polar glaciers once again melted. During the mid-Miocene, about 15 million years ago, the Earth was about 8 degrees Celsius warmer than today. About 12 million years ago, the earth's climate cooled off enough that continental glaciers could again form in Antarctica.
Life adapts to climate changes like this. We, and the polar bears, are here on Earth today as living evidence of the fact that our ancestors lived through these vast temperature swings.
This image was found on http://en.wikipedia.org/wiki/Image:65_Myr_Climate_Change.png.
Now let's look at more detailed temperature data in the more recent prehistoric past. The following graph shows the data from ice cores taken from two Antarctic locations which provide a temperature record of at least the last 450,000 years. The graphs clearly show that temperatures (in the Antarctic) were 3 to 6 degrees warmer than today during the last interglacial period 125,000 years ago.
This image was found at .http://en.wikipedia.org/wiki/Image:Ice_Age_Temperature.png
The following graph zooms in even tighter on the most recent prehistoric times. We can see that the Earth's temperature today is about the same as it was about 5000 years ago when human civilization was beginning. The temperature value shown for 2004 is a one-year average. Please note that because of the limitations of data sampling the data shown by the colored lines has resolution no better than 100 to 300 years. Over time spans like this, one would expect some years to be significantly warmer than the measured average, and other years to be significantly cooler, but we can see only the long-term average when we look so far back into time using current proxy methods.
Some people call the time period shown by the gray bar the "Holocene Optimum", as this was a time when conditions on Earth were close to ideal for life. For instance, the Sahara was not a desert at this time, but was instead a wooded savannah which was ideal for our ancestors to hunt and gather food. This was a time when temperatures were (on average) warmer than the past 50 years.
The idea that historic pre-industrial temperatures represented a "normal" climate is a misconception. In actuality, those so-called "normal" temperatures were significantly below average for a typical interglacial warm period. It is argued that the famous "Hockey Stick" temperature graph probably represents a gradual recovery to a true "normal" temperature for an interglacial period.
This image was found at http://en.wikipedia.org/wiki/Image:Holocene_Temperature_Variations.png.
The following illustration shows the extent of the eastern Great Lakes about 8500 years ago on the left and the temperature and humidity of the Great Lakes region of North America for the past 13500 years on the right. From this it is clear that temperatures 5000 years ago were warmer than today (dashed horizontal line). Note that generally the pattern of humidity mimics the temperature trend. Importantly, there is a period labeled "Warm/Dry" in which temperatures increased faster than humidity. Note also that temperatures in the last 500 years reached a low during the Little Ice Age and have been returning to more typical temperatures since then. "More typical" temperatures, from a geologic perspective, are higher than what modern people would consider "normal" (that dashed horizontal line)!
This image was found in Science News issue 0f 29 August 2009. The graph is from Lewis et al, EOS, 2008.
Modern thermometer records clearly show that the Earth's temperature behavior is in fact quite complicated, even over the span of just one century. The following graphs show the detailed continental U.S. average monthly temperatures since 1895 grouped by season. The most prominent feature of these graphs is the way average temperature moves in cycles about 60 years long. For example, in the winter graph (upper left) notice the gradual increase in temperature from the late 1800's to 1950 followed by a rapid drop in the 1960s and another gradual climb peaking higher than the earlier cycle in 1998. In contrast, summer temperatures (lower right graph) peaked during the Dust Bowl years in about 1935, dropped to a minimum in about 1965, and have been slowly rising since then but are still below the 1930s Dust Bowl maximum. In general, the winter and spring graphs show that temperatures during the most recent cycle were higher than the previous cycle, but the summer and fall graphs show that the earlier cycle centered on the late 1930s had higher temperatures.
Our view of the past is skewed by the availability of reliable data. Many modern data sets like satellite measurements start in the 1970s, a time when world temperatures were lower than average. Presenting conclusions based on changes from such an anomalous "baseline" minimum in the 1970s is misleading, yet very recent (2007!) scientific papers about global warming do exactly this without so much as a footnote about the 1970s "beginning" conditions being demonstrably below average.
Notice that every season during the 1970s had lower-than-average temperatures. It would be expected that sea ice area increased between the 1940s and the 1970s, based on the temperature information in these graphs, but by historic happenstance there were no satellites in orbit to photograph the arctic ice until the 1970s. Recent observations about arctic sea ice disappearing could easily be a normal part of these 60-year temperature cycles, but we've never before observed one of these cycles from orbit, and it just happens that the part we've seen is the upward portion of the cycle.
However, even collecting accurate temperature data poses problems, and opportunities for willful mischief, as the 2009 "ClimateGate" scandal has highlighted (see D'Aleo 2010 http://icecap.us/images/uploads/NOAAroleinclimategate.pdf). Examination of the computer programs used in "global warming" climate studies reveals that temperature measuring stations with intuitively cooler temperatures have been systematically filtered out of the data for recent years but were kept in the data for earlier years (removal of northern sites and high elevation sites in favor of southern and low altitude sites). This results in an apparent upward shift in temperature, but that appearance is mostly the result of removal of lower temperatures from the temperature mapping algorithms. The data kept in recent years also appears biased toward measuring sites with non-representative locations, like near pavement or industrial heat sources like air conditioner heat exchangers (urban heat island effect). In light of these revelations, a complete re-analysis of the raw temperature data by researchers who have not been previously involved is being demanded.
Is the Earth's present-day sea level at the maximum possible level? No, sea level will go higher if more polar glaciers melt. However, compared to the last Ice Age, there are very few continental ice sheets remaining to be melted, and as a result the rate of sea level rise has been gradually decreasing through time. During the last Ice Age, up until about 20,000 years ago when global temperatures started to rise, all of Canada and most of the northern United States and northern Europe were buried kilometers deep with glacial ice. That ice, of course, has long since melted, having run back into the sea thousands of years ago. These days, virtually every ounce of water that falls as snow in the winter gets melted by the end of summer, except on a few rare mountaintops and shadowed valleys.
When world temperatures started to rise at the end of the last Ice Age, sea level rose extremely rapidly at first (130 cm/century), but slowed considerably about 7,000 years ago to 50 cm/century. Between 5,500 years ago and 3,200 years ago, sea level was rising at only 23 centimeters per century. Between 3,200 years ago and today, sea level rise slowed again to an average of about 4 cm/century. Since the 1930's Dust Bowl era, sea level has been in a rising spurt going at about 23 cm/century again, but there is strong evidence that indicates spurts like this are not unusual. Data from south Florida shows that during the past 3,200 years there have been at least 2 episodes of sea level drops and subsequent rises of more than one meter magnitude, so it is likely that the observed late Holocene 4 cm/century rate is a long-term average of higher frequency rises and falls.
First Graph: Tide gauge data for four US East Coast cities from Psulty and Collins,1986. Second Graph: Sea level data for several other US coastal cities from the National Ocean Service. Sea level rates less than the 2.3 mm/yr global average are due to rising land elevations as the Earth's crust continues to rebound from being warped down by the weight of glaciers during the last Ice Age or due to rising by tectonic uplift (Sitka AK, San Francisco CA). Sea level rates higher than 2.3 mm/yr occur in cities where natural soft-sediment compaction leads to land subsidence (notably Galveston TX).
Satellites have been orbiting the Earth and monitoring the polar ice caps since the late 1970's. The following graph shows the northern polar ice cap extents (in millions of square kilometers) over the full history of such data collection. Note that the area covered by sea ice has a very strong annual cycle, and that the subtle effects of the 11-year sunspot cycles can also be detected. Over this history of satellite reconnaissance, both the winter maximum (upper blue) and summer minimum (red) sea ice areas have gradually been decreasing in response to warming of the Earth. Note that the lowest winter maximum sea ice extent occurred in the winter of 2005-2006 and since that time the winter sea ice extents have been increasing. Please remember this 2005 minimum date because it's significance will be explained later in this article (it may have nothing to do with CO2 levels and everything to do with the sun's energy output cycles). Note also that the lowest summer minimum sea ice extent occurred in the summer of 2007 and since that time has been increasing in area. This observed time lag between lowest winter and lowest summer ice extents is (theoretically) because first-year sea ice is thinner and therefore more easily melted the following summer than older and thicker sea ice. I prepared this graph based on data downloaded from the University of Colorado FTP site on 10July2010.
Recent analysis of radar altimeter data from two satellites operating between 1992 and 2003 has found that the central Antarctic ice sheet is thickening by an average of about 1.8 cm per year (Davis et al, Science, June 24 2005), even though temperatures have been somewhat elevated throughout the past decade. This amount of new ice, estimated to be 45 billion tons per year, is equivalent to a drop in the world sea level of 0.12 millimeters per year. Computer climate models indicate that the Antarctic ice cap thickening is the result of higher yearly snowfall due to atmospheric warming. Since warmer air holds more moisture than colder air, precipitation can be increased in certain areas while other areas are drier.
The same thickening of ice pack has been observed in the interior of Greenland, as well (see following map). At the same time, glacier ice along the coastlines of Greenland are receding, as are many glaciers around the world. We must always bear in mind that climate is cyclic, and the worlds glaciers have been advancing and retreating throughout geologic history. Since the end of the last ice age, glaciers have been retreating far more often than they have advanced. Given this history, no one should be surprised if the long-term retreat of glaciers continues into the future (with occasional, anomalous advances). Because satellites were not in orbit during the 1930's when the global climate cycle was in the same phase as today, we do not know whether similar melting happened at that time.
One occasionally hears complaints that the glaciers in the North American Rocky Mountains are disappearing, a "tragic and avoidable loss to future generations". The fact is that during the "Holocene Optimum" world temperatures were significantly higher than today and glaciologists have learned that some of those lamented glaciers did not exist 5000 years ago. The glaciers we see today were rebuilt during the Little Ice Age that is currently coming to an end. The following graph shows how glaciers have advanced and retreated since 1700.
Clearly, glacier advance and retreat is not a steady-state system. It exhibits a cyclic pattern over time, with some decades being dominated by glacier growth (blue) and other decades being dominated by glacier retreat (red). Since the end of the last ice age, glaciers have been retreating more than they advance.
At times during Earth's long history, continental drift has arranged the land masses into very different configurations from those of today. When there were large amounts of continental crust near the poles, the rock record shows unusually low sea levels during ice age climatic cycles, because there was lots of polar land mass upon which snow and ice could accumulate. During times when the land masses clustered around the equator, climatic cycles had much less effect on sea level. Geologists refer to our current world climate as "IceHouse" conditions because of the positions of Antarctica and Greenland near the poles. During the Mesozoic, the world's climate was classified as "GreenHouse" conditions because there was very little land near the poles on which glaciers could grow.
Today's major continental ice accumulations are on Greenland and Antarctica. There is still a lot of water tied up in those glaciers (enough to raise sea level by 67 meters if it all melted, see http://www.grida.no/climate/ipcc_tar) but due to their extremely high latitudes, only Greenland's portion (at most 7 meters' worth) is ever likely to melt to any appreciable extent (again see http://www.grida.no/climate/ipcc_tar). The world's remaining water contained in glaciers and grounded ice caps (outside of Antarctica and Greenland) is equivalent to about 1/2 meter (about a foot and a half) of global sea level rise. Today, sea level is rising about 2.3 mm per year (to keep this in perspective, human fingernails grow approximately 36 mm per year), and only about 20% of that rise is due to water coming from melting glaciers. The remainder of the rise is due to thermal expansion of the existing seawater (remember, global temperatures are rising). It is important to keep in mind that the supply of glacial ice is not infinite, so this rate of meltwater contribution cannot continue indefinitely.
The rest of Earth's permanent ice cover is already floating in the sea (for example, most of the Earth's northern "Ice Cap" is floating). Unlike continental ice, floating sea ice changes sea level only a small amount (about 3% of the ice volume) when it melts. If you doubt this, you can test it for yourself in the kitchen. Fill a glass with water about 3/4 full. Now add ice cubes until the water level is just below the rim of the glass. You'll notice that the tops of some ice cubes are actually above the rim of the glass at this point. Now wait a couple of hours for the ice to completely melt. When the ice is gone, you will find that the water level is still exactly just below the rim of the glass and not one drop has spilled out. This illustrates Archimedes' Principle which states that floating ice displaces it's own exact weight of water. Due to the differences in the salinity of sea ice and the salinity of the seawater it displaces, there is approximately a 3% volume per unit weight difference, which means that about 3% of the melted ice volume will contribute to sea level rise. However, the disappearance of floating ice can have a domino effect on the arctic ocean temperature because white ice reflects 90% more sunlight energy back into space than open water does. This phenomenon is an example of "positive feedback", where a trend is self-reinforcing.
OK, so if Earth's temperature has been rising since long before people built the first cities or started using fossil fuels, what was causing the warming? The short answer is "orbital mechanics and solar output variation". The following cartoon illustrates the most important of these controls on solar radiation reaching Earth's surface. The lower blue and yellow graph shows the predicted summer solar energy (at 65 degrees north latitude) and the amount of relative ice volume on Earth. The graph covers only the past 750 thousand years, but I have personally seen the effects of these same cycles in rocks over 400 million years old. Please note that today's date is located on the right side of this graph (0 years ago).
The close-up of the Earth in the upper left corner of this cartoon mistakenly shows the direction of precession to be counter-clockwise. It is actually clockwise across the constellations, but it looks like it's counter-clockwise when you're standing on the ground looking up (imagine looking outward from the face of a clock). Today, we are in a relatively low eccentricity (nearly circular) orbit. For a graph of the Earth's orbital eccentricity through time, click:http://www.museum.state.il.us/exhibits/ice_ages/eccentricity_graph.html
The Earth's orbit around the sun is warped over time by the gravitational influence of the larger planets (Jupiter, Saturn, Uranus, Neptune). Depending on their changing alignments, the planets' combined gravitational pull gradually elongates the Earth's elliptical orbit around the Sun, and later exerts forces that tend to make the orbit more circular. This effect changes in a precise mathematically predictable way , with approximately a 100,000 year cycle. Not coincidentally, you will notice in above chart that the amount of Earth's ice volume (blue area) has also hit a low about every 100,000 years. We are currently living in one of the most severe low ice portions of the cycle (date "zero" on the blue and yellow chart's horizontal axis), and the Earth would be in that part of the cycle whether or not humans were present on this planet, or whether we were running our civilization on fossil fuel or buffalo chips.
Notice also on the cartoon above that the 100,000 year cycles form a kind of asymmetric saw-tooth pattern in the blue area. There are also smaller "teeth" at about a 21,000 year cycle superimposed on that shape (different researchers calculate the frequencies of those smaller peaks and troughs at between 21,000 and 23,000 years). Those smaller peaks and troughs roughly correspond to the highs and lows of the yellow curve, which tracks the predicted amount of noontime solar energy falling on a square meter of the Earth's surface at 65 degrees north latitude during northern hemisphere summer. The value varies between 400 and 500 Watts per square meter, a whopping 25% change in solar energy delivery at the latitude where ice age glaciers form. This 21,000 year cycle is due to the wobble of the Earth's rotational axis (axial precession), kind of like a kid's toy top will slowly wobble while spinning at a high speed. The effects of precession and orbital eccentricity add together to form a complex pattern of closely spaced highs and lows superimposed on a longer pattern of highs and lows. There is also a variation in the tilt of the axis on a 41,000 year cycle that adds significantly to the pattern, making every second precession cycle stronger than the one in between.
The following cartoon shows the effects of axial precession on the timing of seasons through one 21,000-year precessional cycle, looking down at the Earth's orbit from the north.
This figure is my own work.
In 10,000 years, when spring and summer occur at closest approach, fall and winter will occur at aphelion and will be longer. Again, this effect is most dramatic during times of high orbital eccentricity, so the orbital speed effect is minor today. 10,000 years in the future, people in the northern hemisphere should expect to see longer winters with slightly cooler temperatures and to expect longer and slightly warmer springs and summers, based solely on natural changes in where the North Pole is pointing.
The seasonal duration analysis is my own work based on Naval Observatory data.
Isaac Newton gets the credit for figuring out the mechanics of orbital motion.
Today, Earth's closest approach to the sun (perihelion, 91.4 million miles) occurs on January 4th (southern hemisphere summer) and the farthest point from the sun (aphelion, 94.5 million miles) happens on July 5th. Due to the slow but steady precession of the rotational axis of the Earth, the date of perihelion (closest approach) gradually moves one day later each 58 years, so about 9,800 years from now perihelion will come during northern hemisphere summer. But don't expect a huge change in your local temperature, because the Earth will still be in a low-eccentricity part of the orbital cycle for the next 25,000 years, so the coincidence of perihelion with northern hemisphere summer won't be that different from northern hemisphere summer at aphelion (by definition, low eccentricity means a nearly circular orbit). In other words, during low eccentricity periods we have relatively mild seasons because the Earth is about the same distance from the sun throughout the year. The seasonal extremes are much more pronounced when the Earth is in a high eccentricity part of the cycle.
Why does the tilting and precession of Earth's rotational axis make any difference to world sea level? The Earth's axial orientation controls whether the Earth's northern hemisphere is getting its summer during Earth's closest or farthest approach to the sun in its elliptical orbit. This matters because there is more continental area in the northern hemisphere, and so during colder eras the winter snowpack has more of a chance to pile up on continental bedrock in the north. If (as is the situation today) the closest approach to the sun occurs close in time to northern hemisphere winter solstice (currently perihelion occurs on January 4th in 2004 just a few days after the winter solstice), winters in the northern hemisphere will be warmer than average. As the years go by, winter will get even warmer as precession brings perihelion (and some extra solar energy) during midwinter. Early February is usually the coldest time of the year, but 1566 years from now the earth will be closest to the sun on the first of February and that extra solar energy will make Februaries measurably warmer than today. Now recall the earlier graphs showing continental U.S. average seasonal temperatures: Modern temperature readings show that most of the warming we observe today is occurring during deep winter and spring. Winters and springs are becoming less cold, but summers and falls are cooling somewhat, just as the Milankovitch theory predicts. However, Milankovitch theory predicts a lower rate of temperature change than we are currently observing, and leading global warming theorists blame greenhouse gases for the difference.
It is perhaps counterintuitive, but the southern hemisphere getting warmer (or colder) winter weather doesn't have much effect on sea level because there's not much continental mass in the far south on which glacial ice can accumulate (just the narrow tips of Africa and South America plus Antarctica). No matter how much year-round Antarctic floating ice forms during extra-cold winters or is melted during extra-warm summers, the presence or absence of ice caps floating in the ocean has very little effect on the world's sea level (remember the experiment with the glass of ice and water). The recent breaking off of Antarctic "Ice Shelves" has had absolutely no effect on sea level because they were already floating.
Today the Earth's orbital eccentricity is less than 0.02, but every 100,000 years it stretches out and approaches 0.05 . As the eccentricity of the orbit increases, the axial precession effect on seasonal temperature is expected to be amplified because the sun goes both farther and closer to the sun during the year. Also, as mentioned earlier in the precession diagram's caption, the speed of the Earth in its orbit varies enough to shorten and lengthen the seasons during times of high eccentricity. This orbital cycle correlates strongly with the really big seasonal temperature fluctuations occurring in groups every 100,000 years.
The following graph shows the amount of summertime solar energy falling on four different latitudes through time. Please note that today's date is located on the left side of this graph (0 years ago).
Please pay particular attention to the patterns of amplitude changes through time. At times, the insolation energy difference due to axial tilt, axial precession, and orbital eccentricity is as much as 25%, changing from 400 to 500 Watts per square meter! In the IPCC Fourth Assessment Report (click), the impact of greenhouse gases on the climate is calculated to be between 2.07 to 2.53 Watts per square meter, a mere 2.5% of the natural documented variation in solar insolation due to orbital mechanics.
Today, we're in a relatively low amplitude part of the combined Precession-Obliquity-Eccentricity cycles, where we expect only about a 4% insolation variation across a cycle. Notice especially the pattern leading up to the amplitude minimum about 400,000 years ago, which shows approximately the same pattern as events leading up to present day. And prior to 400,000 years ago we had an extended interglacial ("between ice ages") warm period, though eventually an Ice Age was initiated as a result of a northern hemisphere energy minimum (blue line dips below 420 watts/square meter) that occurred almost exactly 400,000 years ago.
(http://jlevine.lbl.gov/). The glacial data is from Lisiecki and Raymo (2005) and gray bars indicate interglacial periods, defined here as deviations in the 5 kyr average of at least 0.8 standard deviations above the mean. Note that in the last million years there have been at least 10 major interglacial periods, and that the maximum temperature of each has been rising (actually, the amplitude of the difference between max and min temperatures has been increasing).
The following chart shows the solar energy predicted into the future (future is to the right of zero) based on orbital data. Notice that we are currently in a minimum amplitude period (seems to universally cause long interglacial conditions), and that for the next 250,000 years the amplitudes will be increasing (which will cause more contrast between summer and winter temperatures, which seems to be conducive to glacial conditions). 400,000 years from now, we will be back to approximately the same position in the cycle as today, just as we were 400,000 years ago.
The author (Hollan) chose a somewhat higher solar energy constant than Berger and Loutre, so he shows slightly higher energy influx. This difference in assumed solar "constants" highlights the fact that solar output is not actually constant. The sun's energy output changes on a roughly 11-year cycle, being affected by things such as sunspots and other solar disturbances. As a result, record high temperatures tend to get set every 11 years, and record low temperatures tend to get set in the years halfway between!
This chart can be found in: http://amper.ped.muni.cz/gw/articles/html.format/orb_forc.html
The sun itself is not constant in its energy output. Up to now I have discussed only the variations in solar energy caused by changes in the Earth's orbit and inclination. However, there is abundant evidence that the sun itself varies in its energy output. Just as turning up the flame on a stove affects the temperature of the pot above the flame, increasing the energy output of the sun has measurable impact on the Earth's temperature.
Solar cycles (boxed quote is excerpted from http://en.wikipedia.org/wiki/Solar_variation)
Solar cycles are cyclic changes in behavior of the Sun. Many possible patterns have been noticed.
Changes in total irradiance
Changes in ultraviolet irradiance
A proxy study estimates that UV increased by 3% since the Maunder Minimum.
Changes in the solar wind and the Sun's magnetic flux
(above boxed quote is excerpted from http://en.wikipedia.org/wiki/Solar_variation accessed June 2005)
The following graph shows the Carbon-14 signal for the last twelve hundred years. There are several periods of unusually high and unusually low levels shown on this image. This graph suggests that the Earth is today at or near the maximum point in an 800-year climate cycle.
High Carbon-14 levels correlate with high solar energy output.
Low global temperature events are associated with widespread drought and hardship. For example, the failure of the Anasazi civilization in the American southwest corresponds to a low-temperature Wolf Minimum event around 1350 AD, and appears to have been the result of drought and crop failures. At the same time, the harbors of the Viking settlements in Greenland became frozen year-round, and without supplies from Europe the settlements failed. Around 1350 AD in the South Pacific on Easter Island, the people who erected the massive stone effigies disappeared, possibly due to food supply failure as the island became a desert.
The following graphs show frequency analyses of the sunspot cycles and Carbon-14 data. While the sun's oscillations are not as perfectly uniform as a metronome, the cycles fall into fairly tight frequency bands. Exact reconstructions of past and future solar output levels suffer from the slight irregularities of the cycle durations. However, these reconstructions match the historic thermometer and pre-instrumentation temperature proxy data remarkably well, and can therefore be considered trustworthy indicators of future solar energy output.
Note also the 2700 year spacing between the 5400 year solar maximum and the 2700 year solar maximum. If this periodicity was extended to the present day, it would predict that today we live in a time of high solar activity, an interpretation that is supported by other solar-related isotopes, like Beryllium-10.
A spectral analysis of the above radiocarbon data highlights the periodicity of solar activity. Note how clearly the named solar cycles emerge: 210 years (Suess cycle) and 2,400 years (Hallstatt cycle). Note also that there are "side lobes" (arrows) on the major cycles that evidence the slight irregularities of the solar cycles. So far, we do not have sufficient knowledge to predict whether any given solar cycle will have the period of maximum likelihood or be one of the side lobes. Even the periodicity of sunspots has some variation from the 11-year average (between ~10.5 and ~12.1 years).
John Gribbin, in his book "The Case of the Missing Neutrinos (and other curious phenomena of the Universe)" describes the long history of measurements of the Sun's diameter. There are numerous methods for measuring the sun's diameter, and for a long time researchers thought the methods were flawed because they gave inconsistent results. However, when the measurements are graphed over time, they fit into a cyclic pattern with a wavelength of approximately 70-80 years. This is very close to the Gleissberg cycle mentioned above, and could be the explanation for the larger variations in global temperature during the 20th century, which appear to have an average cycle period of about 61 years.
There is also evidence that the 100,000 year cycle in climate is the result of variations in the sun's magnetic field (see 100,000-Year Climate Pattern Linked To Sun's Magnetic Cycles).
The image below shows Carbon-14 production in the atmosphere (lower curve) and a solar output model based on the solar cycles (from Perry and Hsu, 2000). The authors note that archeological evidence shows that times of low temperatures ("M", "T2", "N1", "N3") correspond to historic hardships like widespread drought, while higher "optimum" temperatures correspond to times in which cultural and technological advances, establishment of civilizations, and human expansion were greatest.
Perry and Hsu, 2000.
Being a natural object, the sun does not oscillate with perfectly regular cycle periods, so using mathematically perfect models does not result in an exact match to the proxy data. Nevertheless, the match is good enough that it suggests this method can be used to predict conditions a few centuries into the future with some confidence. Note that Perry and Hsu's model shows that solar output at 2000 AD is at or near a maximum, and that solar output will fall in the near future.
This prediction is supported by global temperature measurements that show that temperatures peaked in 2004 and have been going down for the past 4 years. (Hadley Meteorological Center data accessed July 2008 from link)
Zooming in on more recent history, the following graph shows the correlation between Beryllium-10 in ice cores and sunspot activity since the early 1400s. Note that Be-10 goes DOWN when sunspot activity goes up. Note also the 60- to 80-year cyclicity evident in the curves.
This figure shows two different proxies of solar activity during the last several hundred years. Today's date is on the right side of the graph. In red is shown the Group Sunspot Number (Rg) as reconstructed from historical observations by Hoyt and Schatten (1998a, 1998b) (http://www.ngdc.noaa.gov/stp/SOLAR/ftpsunspotnumber.html#american). In blue is shown the beryllium-10 concentration (104 atoms/(gram of ice)) as measured in an annually layered ice core from Dye-3, Greenland (Beer et al. 1994). Beryllium-10 is produced when high-energy cosmic rays bombard our atmosphere. http://www.lpi.usra.edu/meetings/metchron2007/pdf/4037.pdf
There is an inverse relationship between solar activity and Be-10 production because a more active sun creates a stronger magnetic field, which deflects cosmic rays away from the Earth's atmosphere. ("Variations in Solar Magnetic Activity and the Sun-Climate Connection")
Note the strong resemblance of the beryllium curve and the famed "hockey stick" global temperature graph. It is reasonable to suggest that the temperature changes observed over the past few centuries (and especially the 20th century) are due solely to natural fluctuations in solar output.
The following schematic drawing shows the mechanisms of Beryllium-10 and Carbon-14 production in the atmosphere. Be-10 is produced by high-energy charged particles (cosmic "rays") hitting the Earth's atmosphere at speeds near the speed of light. The paths of charged particles are deflected by magnetic fields, so when the sun's magnetic field is very strong, like during a high sunspot cycle, a smaller fraction of cosmic rays hit the Earth's upper atmosphere.
Carbon 14 is formed by high-energy, high velocity neutrons created in the sun hitting nitrogen atoms in the upper atmosphere. Because neutrons are uncharged, they are not deflected by magnetic fields. Higher neutron production in the sun results directly in more Carbon-14 in Earth's atmosphere.
Beryllium-10 is produced when high energy "cosmic" radiation originating outside the solar system reacts with Oxygen and Nitrogen in the atmosphere. Carbon-14 is created when high-energy solar neutrons react with Nitrogen in the Earth's atmosphere.
The sun's activity level "modulates" the production of these isotopes. An active (stronger) solar magnetic field reduces the number of cosmic "rays" that impact the Earth's atmosphere. Higher production of neutrons inside the sun during active phases increases the production of Carbon-14 in the atmosphere.
Higher levels of solar activity results in lower concentrations of Be-10 on Earth and results in higher levels of Carbon-14 on Earth.
There is a remarkable match between Earth's temperature and proxies for solar irradiance, as seen in the following graph.
The following graph shows sunspot activity observations starting in 1750 AD and continuing through 2005. The approximately 11-year sunspot cycle is very clearly visible. Notice also that three or four cycles in a row will have relatively many sunspots at the peak, and the following three or four cycles will have relatively few sunspots at the peaks, producing a sine wave with a wavelength between 66 and 88 years. Notice the extremely low number of sunspots in 1812, the year Napoleon's army was devastated by an unusually severe winter. Notice also the extremely low values in 1776, when George Washington feared losing his army to desertion due to a bitterly cold winter, and while crossing the Delaware River at Christmas encountered a river choked with ice, a previously unknown occurrence.
Below is a graph of the most recent sunspot activity cycle. It shows that the sunspot activity has remained very low for several years longer than a normal 11-year sunspot cycle would predict. Note that sunspot activity crested in 2001 and was at its lowest in mid-2008, after which activity has slowly increased.
This sunspot curve from 2010 shows that the number of sunspots did not begin to rise in 2007 as the 11-year sunspot cycle would predict. However, this did not occur. The sun has entered a very quiet period that has resulted in record cold temperatures worldwide for a couple of years.
graph source (July 2010):http://solarscience.msfc.nasa.gov/SunspotCycle.shtml
Now look at the solar forcing data that was used in NASA's GISS climate model in the following graph. They have clearly captured the 11-year sunspot cycle, but notice that it slowly falls in the late 1800s and then steadily rises throughout the 20th century, then levels off at this higher level. Knowledgeable critics of global circulation models correctly observe that this critical solar forcing input to the climate model does not display a cyclic pattern into the future as is suggested by astronomic solar research. Astronomers know that the sun's output varies on several superimposed cycles, yet the future projections of this graph do not show this cyclicity. The sun's output does not in actuality go up and stay up. It should rise and fall with a period of about 88 years (the Gleissberg cycle) superimposed on an approximately 200-year period Suess cycle. Remember, this is the sun's energy output they're graphing, not it's effect on global climate. Computer programmers have a phrase to describe this kind of input error and its effect on calculations: Garbage in, garbage out.
Solar output does not in actuality follow this pattern, which was used as input to NASA'a GISS climate simulation. Solar scientists expect solar irradiance to enter a falling stage around 2000 AD (peak of Gleissberg cycle). Such obvious errors as this have provided critics of climate modeling with powerful arguments against the predictions of the climate simulation models.
The figure below is a reconstruction of solar activity over the past 11,400 years. Please note that today's date is zero on the left side. A high sunspot activity correlates with higher solar energy output. Note the maximum solar activity between 4000 and 5000 years ago which correspond to the "Holocene Maximum" (also called "Holocene Optimum"). The period of the longest solar cycle visible in this sample is about 7000-8000 years. It is interesting that 7000 years ago the Earth was at the lowest part of this particular cycle, and after that the sunspot activity increased (along with world temperatures). In the recent centuries the sunspot number has again been approaching the lowest possible levels (probably the cause of the Little Ice Age) and today may be rising just as it did 7000 years ago. If sunspot activity is responsible for the bulk of today's observed global warming, there's very little we humans can do about it.
Reconstruction of sunspot activity over the past 11,400 years. Today's date is on the left side. Values during the most recent century are not shown. Note the approximately 7000-year cyclicity visible in this plot. Sunspot activity began to rise from a minimum about 7000 years ago, and since high sunspot activity correlates with higher solar energy output, Earth's temperature also started to rise at the same time. Note the maximum solar activity between 4000 and 5000 years ago which correspond to the "Holocene Maximum" (also called "Holocene Optimum"). Note also that during the time of the Little Ice Age (500 years ago to today) the sunspot activity has been unusually low. This image was copied from the GNU public copyright site http://en.wikipedia.org/wiki/Image:Sunspots_11000_years.jpg
What is the Earth's Temperature History and how does the Greenhouse Gas CO2 fit into the Global Warming Story? The top 5 warmest years of the past century have all occurred in the past decade.
Overall, the temperature since 1965 has been rising, and a longer-term rising trend started in 1910.
However, this does not tell us everything we need to know about the temperature history of the Earth. We need to look a very long way into the past to see the full range of temperature variability that the Earth experiences.
The following graph shows a summary of CO2 concentrations and global temperature since the appearance of fossilized life on Earth about 600 million years ago. The plot clearly shows that average global temperatures (blue line) today are about as low as they have ever been (12 degrees C), and also that CO2 levels (black line) are about as low as they have ever been.
There are several mechanisms at work removing CO2 from the atmosphere. The most familiar mechanism is the conversion of CO2 into plant material via photosynthesis. This is the mechanism responsible for the dramatic drop in CO2 during the Carboniferous Period about 360 million years ago, during which most of today's coal deposits were formed. Today, however, plants only store carbon temporarily. When plants are eaten or die and decay, their carbon is released back into the environment. Only if plant material is somehow protected from decay can the carbon be permanently sequestered by plants. Ancient coal deposits are the result of protection by burial.
Another little-known but massive CO2 removal process is the building of calcium carbonate sea shells and coral by marine animals which get buried and preserved as limestone. Calcium carbonate is 40% CO2 by weight. Limestone is hundreds of times more common than coal in the Earth's crust, and is forming today around the world wherever marine shellfish or coral exist. As sea water temperatures increase, the tropical habitats of coral reefs will expand away from the equator into regions that today are too cold to support them. Corals and shellfish quickly colonize new habitats because their young progeny float freely on the ocean currents, finally settling and growing anywhere they can find a foothold.
There are also plants and algae that build hard parts from calcium carbonate. The calcareous algae Penicillus and Halimeda are two examples famous among modern carbonate scientists. Halimeda algae have been found to be the primary carbonate sediment producing organism of the tropical seas. A study on the Great Barrier Reef (Drew 1983) showed that huge meadows of Halimeda produced up to 2 kg calcium carbonate per square meter every year. A patch of Halimeda the size of an average living room produces more than 100 pounds of calcium carbonate in a single year under the proper conditions (adequate supply of calcium ions and habitably warm water temperatures).
The last, and by far the largest biologic carbon sink are marine phytoplankton. These microscopic plants grow in massive blooms and eddies in the oceans. When they are eaten by predators, their carbon is incorporated into the food chain. However, if they sink to the sea bottom and are buried, their carbon may stay locked up in those sediments for geologic eons. In fact, most petroleum is derived from these buried micro organisms. When we burn these "fossil" fuels, we are releasing solar energy that was stored millions of years ago by photosynthetic organisms as chemical bonds.
Analysis of ancient atmosphere gas bubbles trapped in permanent South Pole ice caps covering the past 400,000 years (see graph below) shows that the levels of greenhouse gases have changed through time almost in synch with global temperature. As you look at this graph, notice the strong 100,000 year cyclicity of the temperature and greenhouse gas pattern across a time span of more than 400 thousand years. Please note that today's date is located on the right side of this graph (0 years ago).
Carbon Dioxide in today's atmosphere is about 350 parts per million (ppm) or 0.035%. That's just barely off the top of this chart and represents an excess of about 25% (350ppm compared to the 275ppm expected from our position on the ice core data cycles). There is little doubt that this excess 25% is the result of human civilization burning fossil fuels.
As you carefully examine this data, please notice that historically the black line (CO2) tends to LAG BEHIND the changes in temperature (red line) during periods of most rapid cooling. Apparently, a high concentration of CO2 does not prevent rapid atmospheric cooling at the onset of an Ice Age. Methane is perhaps more important than CO2, since it is 10 times more effective a greenhouse gas than CO2 and its concentration almost exactly tracks temperature throughout the data record.
It is important to note that despite several huge shifts in global temperature, the Antarctic glaciers that provide this data have not melted during the past 400,000 years, a time period during which world temperatures have occasionally been significantly higher than today. This suggests that there are latitudes above which temperatures never get high enough to melt the polar ice caps, even during the most extreme interglacial warm periods of Earth's global climate cycles.
This graph can be found on the internet at: http://www.grida.no/climate/ipcc_tar/wg1/fig2-22.htm
Just within the last 15,000 years, the temperatures recorded in these ice core samples range from 4 degrees hotter than today to 10 degrees colder than today. Also, the temperature during the current interglacial period (only the past 5,000 years or so) has been bouncing between 2 degrees Celsius hotter and 2 degrees lower than today. During the period known as the "Holocene Temperature Optimum" (around 5000 years ago) the Earth's temperature was about 1.5 degrees C warmer than today. Keeping this data in context, the 1 degree rise in the past 500 years (remember that earlier graph) should be understood as being entirely within the 'noise' during any typical high temperature interglacial period.
Notice also how similar the temperature changes (red line in preceding graph) are to the 100,000 year pattern of global temperature cycles predicted by the Milankovitch orbital cycles. I find it very interesting that the CO2 concentration (black line) remains high during the periods of most rapid temperature decline (at 400, 310, 220, and 110 thousand years). However, the concentration of methane, a naturally produced greenhouse gas, seems to almost exactly track the temperature change.
Please note that the zero on the preceding graph's right vertical axis is today's average temperature...this chart shows that the Earth has been 2-4 degrees warmer than today at least 4 times before there was human civilization! It is not logical to hold human civilization responsible for these four huge increases in greenhouse gases over the last 400,000 years because we know that human civilization was only beginning to emerge a mere 5,000 years ago. Also, remember that humans have been burning fossil fuels and generating "greenhouse gases" like carbon dioxide for only the past 200 years (just a fraction of an inch on the right side of the graph, about half the width of the zero character on the lower right axis).
Scientists need to discover why the concentration of methane follows so closely (yet CO2 lags slightly behind) the coming and going of Ice Ages. Which is the cause, and which is the effect? Did climatic warming (independently driven by orbital cycles) somehow cause the release of greenhouse gases from natural sources like bogs and melting permafrost, or did rising concentrations of greenhouse gases in prehistoric times cause global warming? Scientists have a lot more work to do before we can answer these questions and say we completely understand global warming in a human-free environment.
So how well can we predict global temperature changes? We can do pretty well, over the long term, but only in general trends. We can't do so well on the short term, because there are lots of temporary factors like volcanic eruptions, solar flares, and forest fires that have global effects on weather. Also, the simple presence of clouds, ice, or snow can cause the climate to shift toward colder weather in what is called a feedback mechanism, and these effects are notoriously difficult to represent mathematically in computer climate models.
In general, we should be able to make some broad predictions based solely on the orbital cycles and solar output cycles. For instance, our sun is currently in an active phase, which means its energy output is at a near-maximum right now. This will naturally push Earth's temperatures higher than average, even if there were no human-produced greenhouse gases. But projections of these cycles into the future indicates a gradual decrease in solar output that will ultimately lead to another global ice age.
In about the year 1250 AD each European summer was coming when the Earth was farthest from the sun (the Earth's north pole was tilted toward the sun at aphelion), which led to the northern hemisphere's "Little Ice Age" from which we could still be emerging. We should expect winters and springs to get warmer for the next 5000 years as perihelion progresses into April (spring thaw will come slightly earlier each year). In 9,800 years, when the Earth is at closest approach to the sun during July (just as it was ~11,300 years ago), summers in the northern hemisphere will be hotter than they are now and winters will be colder than they are now. This is what we think caused the last ice age to end about 10,000 years ago: centuries of warmer than average northern summers gradually melted away the northern hemisphere glaciers faster than winter ice could accumulate.
What does any of this have to do with the controversy over Global Warming? Only this...geologists and climatologists are certain that the Earth has gone through periods both warmer and colder than what we call 'normal' today. The planet has gone through these temperature fluctuations on a regular and generally predictable cycle, and there is overwhelming evidence that it has been doing this throughout geologic history.
Based on our current-day position on the Milankovitch orbital cycle chart, we should fully expect the planet to be warming, even if it was entirely devoid of human population. We should also expect to observe the concentration of greenhouse gases in the atmosphere rising to follow the global temperature, just as the ice cores prove has happened many times during geologic history in the complete absence of human influence.
Further, the Milankovitch theory tells us to expect the solar energy delivery to the northern hemisphere during winter and spring to continue to slowly rise for the next 9,800 years, at which time the North American summer solstice will occur at perihelion. In the meantime, we should not be surprised to be experiencing episodes of relict glacier melting, measurable sea level rise, and associated weather changes. We just don't know exactly how many degrees per century of this wholly natural temperature change to expect on the short term, given the fact that world temperature has bounced around so wildly in the past. Until we understand the natural variations, we cannot untangle the human influences from today's temperature measurements.
Geologists and paleoclimatologists know that in the past the Earth's temperature has been substantially warmer than it is today, and that this warming has occurred under purely natural circumstances. Until we can say precisely how much of the current global warming and greenhouse gas increase is the result of this normal temperature cycle, we will not be able to measure how much human activity has added to this natural trend, nor will we be able to predict whether there will be any lasting negative effects. Hundreds of research projects are currently underway to understand climate change, but the answers are not yet known.
The most definitive studies involve the use of computerized global circulation models (GCMs) to mimic past climate and project global climate into the future. To be credible, these GCM experiments must adequately describe past climatic behaviors like El Nino events and seasonal variations. Most of these credible models predict continued global warming, assuming no reductions to greenhouse gas emissions are made. However, GCM-based scientific journal publications linking human greenhouse gas emissions to global climate change usually express their conclusions using phrases like "May be", "Could be", "Suggests", and "Cannot be ruled out" to indicate that a more-than-negligible uncertainty remains in their predictions.
NOAA temperature records for the US suggest that surface temperatures may have crested in 1998 and are starting back down (see following graph, prepared using data through 2004). Since the sunspot activity is predicted to start the upward swing of its cycle in 2006, it should not be surprising if temperatures increase somewhat over the next 5 years before the downward portion of the cycle becomes apparent.
The following chart shows the "best estimate" world temperature averages (blue line) based on thermometer readings since the 1850's when these instruments were put into widespread use and records have been kept. The pink line is the result of statistical analysis of solar cycles (with only a 0.2 degree per century non-solar slope which I presume to be the human-caused component, but could EASILY be the effect of a solar or orbital component with a period greater han 210 years). Note the observed temperature peak in 2005 and how the solar cycle predictions match this. Recall now the Arctic Sea Ice extent graph that showed the minimum sea ise extents occurring in the 2005-2007 timeframe, and gradual ice extent increases since that time. These ice observations are entirely consistent with this graph's world temperature predictions that are based almost entirely on extremely well documented solar energy output cyclicity.
Note that the world temperatures reached a maximum in 2005 and have been falling since that time. This reversal is accurately predicted by my combined solar energy output cycle sine wave equation. The equation predicts that temperatures will fall to "average" temperatures in 2015, rise slightly in the early 2020s, and then continue to fall below "average" until 2034. If the plot was extended further into the future, you would see the equation predicts that temperatures will rise to near current temperatures around 2050 (based entirely on solar energy output cycles with only a very moderate 0.2 degree per century human influence). As with any scientific hypothesis, the predictions of this theory must be tested against real-world observations and be either modified or discarded in light of data that contradicts the predictions.
A similar analysis of solar cycles over a much longer timeframe can be found in http://www.pnas.org/content/97/23/12433.full.pdf
Isn't it interesting that the falling temperature trend of the past several years has not been widely reported in the mass media? Also very interesting is that nobody in the business of selling Global Warming alarmism to the public wants to talk about the summer of 2009, when temperatures were so low that newspapers across the world were calling 2009 the "year without a summer". Colleagues of mine doing field work in the Italian Alps during July 2009 had so much icy weather that they lost several days of productivity because they were forced to stay indoors for their own safety. The winter of 2009 was also unusually severe, with record cold temperatures being set across the northern hemisphere. Several Global Warming rallies had to be cancelled because of record cold temperatures and snow.
This 2009 graph is my own analysis of publicly-available data.
The effects of global warming will force populations to adjust to the new conditions, just as we have done throughout human history. Rising sea levels will flood low lying coastal areas. Countries like the Maldives are already feeling the need to artificially protect their islands from rising waters. However, almost all coral islands exist only because they were recently (geologically speaking) underwater. An example closer to home is the cruise ship harbor of Georgetown, Grand Cayman Island in the Caribbean, which consists of a platform of prehistoric coral reef that is found more than a meter above today's sea level (see photo below). A modest sea level rise of just one meter will put large portions of this island underwater at high tide.
The author standing on a Late Pleistocene age coral reef in Georgetown, Grand Cayman Island in July 2004. The presence of this stranded reef proves that sea level was high enough that those corals were recently under water because these corals could not have grown above low tide. At the time these corals were growing, low tide sea level was above the top of this reef and mean sea level was probably about knee level. The geologic record abounds with this sort of evidence. Geologists know that, just as has happened in the past, sea level will once again rise above these now-dead reefs when the world's temperatures get back up to where they were in the past.
As another example, southern mainland Florida is exposed seafloor. When sea level was higher during the last Pleistocene interglacial period (about 125 thousand years ago), the entire Miami area was under water and coral reefs grew on the plains south and west of the city. The southern coast of Key Largo is made up of exposed coral reefs rising 2 meters above current-day sea level.
This flooding has happened before and will surely happen again. This fact about life on Earth should not come as a surprise, since anybody who looks will find that there are seashells in the soils of most low-lying coastal areas that indicate that they were recently (geologically speaking) underwater. For example, I remember being a toddler in the Los Angeles area, digging holes in our home's backyard several miles from the beach, and discovering new-looking saltwater scallop and clam shells!
The fundamental problem is not that mankind is causing the sea level to rise, since that rise has been going on since before humans invented agriculture. The underlying issue is that modern civilization has built huge cities, vital ports, and expensive resorts in places that were recently under water and will soon again be under water. The familiar Biblical story of Noah dates to an ancient time when sea level was rising catastrophically fast due to the melting of glaciers as the world's temperatures warmed at the end of the last Ice Age. If we could find it, Noah's settlement is today under several hundred feet of sea water in the Gulf of Suez or in the Arabian Gulf (see this link) north of the straits of Hormuz. Anthropologists studying the Ice Age route of migration of native Americans from Asia into Alaska know that the coastal route these immigrants followed is today under 300 feet of sea water.
Another predicted effect of global warming is that tropical storms are expected to increase in number and intensity as a result of increasing ocean temperatures. This again affects those who live in low-lying coastal areas in the subtropical latitudes. Tropical storms feed off of the temperature difference between the sea water and the overlying air. This is why the peak hurricane season occurs in the fall, when the air is cooling off, but the water is still carrying heat left over from summer. The ocean actually takes 2.5 years to gain and lose heat in response to changes in solar output, but the atmosphere responds almost instantly. As a result, the ocean can carry excess heat from the previous year and may spawn severe tropical storms earlier in the hurricane season if the air is cooler than normal during early summer. We may be seeing this happening...Spring 2005 was unusually mild on the Texas Gulf Coast, and yet we got Hurricanes Denise and Emily in July 2005. The powerful 2005 Hurricanes Katrina and Rita gathered their energy from the temperature difference between the warm ocean water and the cooler atmosphere above it.
People wonder if the number or severity of tropical storms is increasing. The plot below shows NOAA's compilation of named tropical storms and hurricanes since 1850. There has been a gradual increase in the total number of named storms, but the number of hurricanes (red and green together) during the last decade are very similar to peak periods in the late 1880s and early 1950s. There is a 60-year cyclic pattern in the number of hurricanes, and that pattern predicts that the Earth's climate should be in the part of the cycle with more than average numbers of tropical storms. The IPCC Fourth Assessment Report (click) concludes that the number and intensity of tropical cyclones is not any different today than it has been in historic times.
Note the 60-year cyclicity in the storm numbers. Based on this pattern alone, we would expect a somewhat larger than normal number of named storms at this time. The Beryllium-10 measurements inversely tied to solar activity levels closely mimic the shape of this plot.
As temperature rises, so does evaporation and therefore rainfall also increases (what goes up must come down), but the distribution of rainfall will not be uniform. Just like today, there will be places in the world where evaporation exceeds rainfall (arid climate) and places where precipitation exceeds evaporation (humid climate). The locations of these specific climates will be different from those of today, and adaptation will be required. Keep in mind, however, that these changes in climate occur naturally (today's Sahara Desert was a temperate woodland during the Pleistocene, and temperate forests once extended to the Arctic Ocean in areas that are treeless permafrost tundra today).
Is there any positive effect of global warming? Agricultural research indicates that a small amount of global warming could actually be beneficial to humanity. CO2 is a plant nutrient, so higher atmospheric CO2 results in more robust crop growth.
Also, warmer marine air means more evaporation from the oceans and therefore more freshwater precipitation on the land (good for some present-day deserts, but bad for flood-prone areas). Historically, times of drought occur during colder than normal climate cycles (see http://www.pnas.org/content/97/23/12433.full.pdf).
Increasing temperatures allow for forests to expand into current-day arctic tundra at the extreme northern and southern latitudes and high altitudes, a process that both releases carbon dioxide locked up in the frozen permafrost soil and later removes carbon dioxide from the atmosphere as carbon-rich wood tissues are grown (a temporary positive and then negative feedback mechanism). For example, during past warm eras (the Holocene Optimum), the timberline was 0.4 kilometers higher in elevation than today (that's why we find ancient dead trees hundreds of meters above today's timberline in North American mountain ranges).
Life is surprisingly adaptable to climate change. Interestingly, fossil remains show that the greatest biologic diversity has occurred during times of warmer than average climate. Life on Earth seems to tolerate warmer weather better than ice and snow, which is just what intuition would suggest. Life on Earth was not wiped out during prehistoric temperature maximum events (when the Earth's temperature was even higher than the worst-case global warming computer model) and the fact that we are here today proves it.
In Al Gore's documentary "An Inconvenient Truth", he claims that polar bears are on the verge of extinction due to increasing arctic temperatures. No one ever mentions that polar bears have had numerous opportunities to go extinct during the past interglacial warm periods, yet they survive today as proof of their resilience to climate change, natural or otherwise. I would not impugn the integrity of these sincere wildlife advocates, but one must admit that communicating the facts about past climate changes does take the wind out of their sails somewhat.
Humanity still has a long way to go in understanding the feedback mechanisms that may affect global climate. For instance, warmer air means more evaporation and therefore more clouds, but daytime clouds tend to reflect sunlight back into space, making temperatures cooler (negative feedback). However, nighttime clouds reflect infrared radiation from the ground back toward the surface, thus keeping nighttime temperatures higher (positive feedback). Glaciers and snow cover reflect much of the incident sunlight back into space, thus reducing temperatures further (positive feedback). But if those glaciers cover the high-latitude forests, the carbon dioxide formerly sequestered by those trees will eventually return to the atmosphere and cause atmospheric warming in a negative feedback loop. As global temperatures cause increased freshwater runoff into the arctic oceans, that less saline seawater will freeze at higher temperatures during the winter and reflect more sunlight back to space (negative feedback). There is a very recent theory that increases in ocean temperature can cause melting of methane-clathrate ice accumulations just below the seafloor, which can add substantially to the atmosphere's methane content, resulting in a positive feedback situation (ocean warming melts clathrate, releasing greenhouse gas methane, which makes it even warmer, melting more clathrate). On the extremely complex subject of human-induced global warming and the associated feedback mechanisms, it is quite accurate to say that we don't yet have all of the answers.
The opinions and interpretations expressed in this web site are solely those of the author and do not reflect the opinions or interpretations of his employer.
Greg Benson is an earth scientist with 34 years of geologic study and 29 years of experience in the oil and gas industry. He earned his degrees in geology at the University of Southern California (1979) and the University of Arizona (1981) and currently works as a research specialist in geologic modeling.
For information about Milankovitch's Astronomical Theory of Climate Change, go to:
Other global warming web pages and information sources: http://www.grida.no/climate/ipcc_tar
and: Solar Climatic Effects (Recent Influence) http://www.co2science.org/subject/s/summaries/solarrecin.htm
and: "The Sun's role in Climate Changes " ( http://zeus.nascom.nasa.gov/~pbrekke/articles/halifax_brekke.pdf ) Proc. of The International Conference on Global Warming and The Next Ice Age, 19-24 August, 2001, Halifax, Nova Scotia
and: Global Annual Average temperature data set TaveGL2v (http://www.cru.uea.ac.uk/cru/data/temperature/)
This web page shows the dates for perihelion and aphelion from 1992 through 2020: http://aa.usno.navy.mil/data/docs/EarthSeasons.html
For a great image of the inner solar system orbits go to: http://ssd.jpl.nasa.gov/orbits_inner.html
100,000-Year Climate Pattern Linked To Sun's Magnetic Cycles http://www.sciencedaily.com/releases/2002/06/020607073439.htm
Sharma, Mukul (2002) "Variations in Solar Magnetic Activity and the Sun-Climate Connection" http://www.dartmouth.edu/~earthsci/labs/sharma/solar.html
"The production of 10Be in the Earth's atmosphere depends on the galactic cosmic ray influx that, in turn, is affected by the solar surface magnetic activity and the geomagnetic dipole strength. Using the estimated changes in 10Be production rate and the geomagnetic field intensity, We have calculated variations in solar activity for the last 200,000 years. Large variations in the solar activity are evident with the Sun experiencing periods of normal, enhanced and suppressed activity. The estimated variations in solar surface magnetic activity are also strongly correlated with the marine d18O record on a 100,000-year timescale. As the marine d18O record reflects changes in the amount of ice locked up on the Poles, We have proposed that variations in solar activity control the 100,000-year glacial-interglacial cycles of Ice Ages. If this hypothesis holds up to further scrutiny, this would open up new avenues of investigations into the Sun-Climate relations. At present, I am working to devise experiments that would prove or disprove this proposition."