Typ 
Figure 1. Arctic Ocean, from www.geology.com/world/arctic-ocean-map
　
ARCTIC ICE
 ---------------    By Larry L. Olson, PhD, PE
　
THE ARCTIC AS A DESERT
Deserts of the earth comprise approximately 10% of the globe’s entire area. They are the primary areas where the excess heat of the earth is radiated into outer space. They are efficient radiators because of their low concentrations of the primary green house gas---water vapor. Typical temperature differences between day time and night time may exceed 25 degrees centigrade. The night time heat is radiated directly to outer space.
The oceans of the earth comprise 71% of the earths surface area and they are the primary heat storage areas of the world. A one degree centigrade temperature rise of the oceans of the world would require 1.413 x 10**24 calories. This is approximately 22 billion times the amount of heat that all of man kind generates each year from the burning of fossil fuels.
If we could connect the oceans to these deserts, we could transfer the heat from the ocean waters to the desert areas and cool the planet. Of course, Nature has been doing this for millions of years.
The arctic ocean is the largest desert on earth, approximately 14.1 million square kilometers. The second largest desert is the Antarctic continent, with an area of approximately 14.0 million square kilometers. The sum total of all of the hot deserts in the world have an area of almost 16 million square kilometers. The definition of a desert is that the area have a total yearly precipitation below a certain level. The Antarctic continent, which is 2/3 the size of North America, definitely is a desert. The Arctic Ocean is currently a desert in the winter, but is less desert-like in the summer because of its increasing humidity.
Both the Arctic and the Antarctic are connected to the oceans of the world--the Arctic more than the Antarctic. Heat is transferred from the equatorial portions of the world to the Arctic via a portion of the Gulf Stream which flows north through the Atlantic ocean and over a underwater ridge between Greenland and Scotland. A smaller flow of ocean water from the Pacific ocean enters the arctic through the Bering Strait between Alaska and Russia. These warm waters (approximately 10 degrees centigrade) flow at the surface into the Arctic ocean, duck under the ice, melt the ice which requires heat from the hot waters, cool to approximately 0 degrees centigrade, dive to the bottom of the Arctic ocean, and then flow out of the Arctic Ocean over the Greenland Scotland ridge, back into the Atlantic ocean. They then flow along near the bottom of the Atlantic ocean south to the Indian ocean where they rise to the surface, are warmed and then start the trip again. This flow of water is also mimicked in the Pacific Ocean, but only occasionally is the very warm water channeled through the Bering Strait to the Arctic Ocean. This system of inter-ocean rivers is called the great ocean conveyor.
　

Figure 2. Great Ocean Conveyor (Atlantic), from www.whoi.edu
　
　
The flow is so large in this conveyor, that a new flow rate term has been established. It is called the “Sverdrup”, abbreviated as Sv, and one Sv is equivalent to one million cubic meters per second (265 million gallons per second). Typical flows in the gulf stream are 30 to 50 Sv in the Florida area and rise to 150 Sv in the Newfoundland area. To put this into perspective, all of the rivers that flow into the Atlantic ocean have a flow rate less than 1 Sv. 
FLOW OF WATER AND HEAT INTO THE ARCTIC OCEAN
The flow of Atlantic Ocean water into the Arctic Ocean has been estimated and measured using various techniques and flows range from 8.5 to 19 Sv. Flows from the Pacific Ocean, through the Bering Strait have been estimated from 1 to 5 Sv. Based on review of several reports, an average value of 12 Sv will be used for the Atlantic portion and 1 Sv for the Pacific portion.
Historically, the Atlantic water has average temperatures of about 8.5 degrees and the Pacific water has averaged 6.5 degrees. During global warming, these temperatures rise, particularly for the top 100 meters of ocean water. Measured values for the Gulf Stream in the Atlantic Ocean and the Alaska Current in the Pacific Ocean show increases of 10 to 15 degrees Centigrade during the late 1990’s and early 2000’s. However, by the time these surface waters reach the arctic Ocean, they are closer to the first temperatures quoted. This is because the warm surface waters mix with the colder surface waters coming out of the Arctic Ocean.
Using these flow rates and temperature values of 8.5 and 6.5 degrees centigrade, we can calculate that the Atlantic Ocean water contributed additional 2.2 x 10**21 calories per year to the Arctic Ocean and the Pacific Ocean water contributed 1.68 x 10**20 calories per year to the arctic Ocean. These figures were derived assuming that the water exiting the arctic Ocean is at zero degrees centigrade. These figures, which total 2.37 x 10**21 calories per year, represent the normal amount of heat that the arctic Ocean/Desert dumps to outer space each year. This is in addition to the summer time heat from the sun that must also be dumped to outer space.
So, the process that goes on all of the time in the Arctic is that during the summer, the heat load from the sunshine plus the heat load from the Atlantic and Pacific Oceans, via the great ocean conveyor, is so large that the arctic area cannot dump all of it to outer space. Consequently, the ice on the surface of the arctic Ocean melts. As the temperature of the arctic increases, the amount of heat that can be dumped to outer space increases (the Stephan-Boltzman equation predicts a fourth order temperature dependence) and reaches a peak heat flow rate in mid summer. As the summer progresses, the amount of radiant energy from the sun decreases until such time that the heat dump rate is equal to the heat input rate from the sun and the great conveyor and no more ice melts. As the season progresses toward autumn, the heat dump rate exceeds the heat input rate and the ocean begins to freeze. As winter comes, the ice thickens, and because the ice is an insulator, the heat dump rate to outer space decreases. When the ice is it’s thickest, the heat dump rate to outer space is at it’s lowest. During this time, the ice is being melted on the bottom because of the great ocean conveyor bringing the equatorial heat into the arctic Ocean from the Atlantic Ocean, there is no heat input from the Pacific Ocean, because the surface of the Bering Strait is frozen which blocks the hot surface waters, and in late winter, the heat transfer rate to outer space is at it’s absolute minimum. The thickness of the ice at any particular location in the Arctic Ocean is a function of how the underwater currents flow: the ice is thicker where there is very little flow from the great conveyor and thinner where there is a lot of the warm water flow. As the earth progresses in its orbit, the sunshine to the arctic increases, and the cycle repeats itself.
It is interesting to note that the 2.37 x 10**21 calories entering the Arctic Ocean via the great ocean conveyor would melt approximately 2.96 million cubic kilometers of ice. Comparing this number to the variation of ice area in Figure 3, it can be seen that if the ice thickness is one meter, then almost all of the melting between the winter and summer is attributable to the great ocean conveyor. That is not logical because the ice thickness is, on the average, more than 1 meter thick. Data presented in www2.sunnySuffolk.edu indicates that the average winter ice thickness was approximately 3 meters in 2004 and then decreased in a linear fashion to 2.5 meters in 2008. This shows that approximately half of the ice melt is caused by the heat from the equator via the great ocean conveyor and the other half comes from other sources. These other sources are increased air temperatures, normal sunlight, variation in albedo due to any number or sources but probably primarily from more open ocean in the summer, and increased fresh water flow because of glacial melt and runoff.
　

Figure 3. arctic Ice Area, from www.appinsys.com/Global Warming/RS_arctic.htm
It would be nice to be able to say that in previous years that the volume of arctic ice came back to the same level each year prior to global warming, but we do not have the data to make that statement. We only started measuring the area of the ice in 1959 and reliable measurements of the ice thickness were not begun until 2004. We have historical evidence that shows that sailing ships accessed some of the arctic Ocean in the 1800s that have not been accessible by surface ships since that time. We further have pictures of submarines surfacing in open water at the north pole in May of 1987.

Figure 4. Submarines At North Pole, from www.john-daly.com/polar/arctic.htm. Note the wind rows
 Of ice that is characteristic of the Arctic Ocean.
　
Evaluation of the ice--no ice picture in the arctic Ocean is further confused by the way the wind acts on the ice. Winds get rather brisk in the arctic in the winter, and it is very common to see the ice jammed together in “wind rows”. When the normal ice thickness is 1 to 2 meters, these windrows can have observable ice thickness above the water of 15 meters or more. Of course, this means that the ice below the water is much thicker. This also means that the open water between the wind rows is larger than normal.
Wind direction and speed also have a marked influence on the transport of ice out of the arctic. When the wind direction is generally from the West and strong, no matter what time of the year, the wind can break up the ice and send it into the Atlantic where it melts. Ask the people on the Titanic about this phenomenon.
The other compounding phenomenon that comes into play is the effect on the ice melt rates when the ice diminishes and more open water is present. Most scientists look only at the summer time phenomenon and say that the open water has a lower reflectivity which causes more solar energy to be absorbed. This causes more ice to melt and increases the overall rate at which ice disappears. They say that there is a “tipping point” at which this process becomes so dominant that the ice will disappear entirely. This author agrees with that analysis, but must point out that you must also consider the “rest of the story” as Paul Harvey would say.
When there is no ice on the arctic Ocean as the winter approaches, there is no insulating barrier between the warm waters of the great ocean conveyor and the air. This means that the heat transfer rate is at its highest and the waters of the arctic Ocean will be cooled rapidly and sink to the bottom. However, because the air is getting lots of energy, it will not be as cold and the pressure difference between the central Atlantic Ocean and the arctic region will not be as high. It has been shown that the rate of transfer of hot water into the arctic Ocean from the Atlantic Ocean via the great ocean conveyor is directly dependent on this atmospheric pressure difference. Therefore, in the absence of ice, the amount of heat coming into the arctic Ocean will diminish and ice formation will be accelerated. So the overall system is somewhat self regulating.
When the ice diminishes, the character of the ice will change. arctic ice is characterized according to whether it is first year ice or multi-year ice. This is rather self explanatory. But its implications are not necessarily so obvious. Multi-year ice means that the total amount of heat coming into the arctic, both from sunshine and from the great ocean conveyor is not large enough to melt all of the ice and so it survives to be combined with the next year’s ice. So, by keeping track of first year ice and multi-year ice and looking at this trend over several years, it is possible to have a measure of the total amount of heat that comes into the arctic. In fact, that is what is being done since 2004 and the trend is that the amount of multi-year ice is diminishing rather rapidly.
Other issues that will effect the extent, volume, and character of arctic ice is the flow of fresh water into the arctic Ocean, the amount of cloud cover variation caused by global warming and the variation of ice reflectivity due to incorporation of dark colored particles into the ice and the effect of the fresh water floating atop the saline water. These have all been studied and each have been given credit for some of the alteration in arctic ice.
So, it is obvious that arctic ice is a rather complex issue---just as is almost every aspect of global warming. It involves flow rates of water, heat content of ocean currents, wind strength and direction, pressure differences between two widely separated points in the atmosphere, solar radiation, reflectivity, insulating effects of ice, melting rates, differences between first year and multi-year ice, atmospheric temperatures, fresh water flow rates, cloud cover, salinity stratification and reflectivity variations with ice contaminants. Add to that the emotional appeal of Polar Bears, and the facts sometimes get overlooked. Such may have been the case in the year 2007.
THE 2007 DISAPPEARING ICE SCARE
arctic ice melting came to everyone’s attention around the world in a dramatic fashion in 2007. This was caused by dramatic headlines in our press and from the comments of one scientist. Both the press and the scientist were correct, because the data showed that the minimum extent of the area of the ice was down by 15% from what would be expected at that time of the year. This left a very large extent of open water---much more than normal.
Because the ice was disappearing at what was presented as an accelerated and abnormal rate, there was much talk about reaching the tipping point and speculation about the disappearance of the polar bears. Of course, it was all blamed on global warming, which was then blamed on the presence of carbon dioxide in the atmosphere, which was then blamed on man’s insatiable appetite for fossil fuels. To some extent, these conclusions were all correct. However, very little attention was given to the normal cycles that control arctic ice and the real reasons for the apparent accelerated melt rate for that year. The following discussion paints a picture of why the ice in the arctic Ocean went down so rapidly in the year 2007.
Some years, the water flow from the Pacific into the arctic is much less than 1 Sv and some years, it is much more. Studies from researchers from the University of Washington has shown that the flow through the Bering Strait is highly dependent on the wind direction. If the wind blows from the North, the flow is primarily to the South, whereas if the wind is from the South, the flow is primarily to the North.
In the year 2007, a high pressure set up over the Gulf of Alaska. High pressure in the northern hemisphere has clockwise flow of air around it when looking down from above. This caused a northerly and northeasterly flow of air in the Bering Sea. The Alaska Coastal Current, which contains the hot water from the North Pacific Current normally flows northerly past the Alaska mainland and then bends to the west and flows past the Aleutian Islands and then turns back south to rejoin the North Pacific Current. With the northeasterly direction of the winds, the Alaska Coastal Current was driven through the Aleutian Islands, hugged the westerly coast of Alaska and then proceeded to flow through the Bering Strait, hugging the eastern shore. Satellite pictures of this area show this very clearly.
　
　
　
　

Figure 5. 2006 Satellite Photo, from www.ak.aoos.org/output/FNMOC/2006. Contrast this to 2007,  shown in the next figure. Note: red is hot and blue is cold.

Figure 6. 2007 Satellite Photo, from www.ak.aoos.org/output/FNMOC/2007. The water in the Arctic
  Ocean got much warmer than pictured in this photo.
　
　
This hot water poured into the arctic Ocean and accelerated the ice melt rate in that area significantly. Some of this hot water got by the measuring stations in the Bering Strait before the researchers caught it, but they deployed a temporary buoy on the very east side to the straight so they were able to measure the temperatures and estimate the flow. Because of the south wind, the flow through the Bering Strait, from the Pacific to the arctic Oceans was significantly higher than the 1 Sv.
This high pressure stayed in place for many days. The winds in the arctic Ocean, to the northeast of Alaska were strong enough that large quantities of ice in the form of ice bergs and growlers was pushed out of the arctic Ocean, through the channels around Greenland. This ice went into the North Atlantic and was melted.
Most of this information was put together months after the fact, but these are all phenomena that would be expected to occur when looking at the Arctic Ocean. In 2008, the ice coverage went back to or slightly above the projected amount. Obviously, this was mostly first year ice because a lot of the multi-year ice had been pushed out of the area. The 2009 ice coverage was down about 8% from that which was expected, and so everyone is watching the 2010 ice very closely.
PROJECTIONS
There is no question that the heat load going into the arctic Ocean is greater than the present capability of the arctic Ocean to dump it to outer space. The decline in the total amount of ice (a loss rate of about 5% per year) signals that the input heat rate exceeds the heat dump rate. However, the amount of ice coverage is still larger than it was in the early 1800s when ships were able to go where ships have not been able to go since that time.
Most temperature graphs that go back prior to 1700 show that there was a general global warming from about 1700 to 1800.

Figure 7. Temperature Graphs, from www.ncdc.noaa.gov/paleo/glovalwarming/images/last2000-large.jpg.
　
　
This would account for the lack of ice in the arctic Ocean in the early 1800s. These same temperature graphs indicate that the amount of temperature rise during this episode was less than half of the current temperature rise. This means that we can expect that there will be less summer time ice in the arctic during this current episode of global warming than there was in the 1700-1800 episode.
Quantitative predictions as to the exact amount of ice loss and when the loss is expected to turn around are predictions beyond the qualitative nature of this article. However, based on the quantitative predictions in another one of the author’s articles, it can be projected that the minimum summer time ice coverage will occur in the years 2016 to 2018. This prediction is based on the variables being quasi-constant, and as this review of ice in the arctic Ocean has shown, about the only thing that is constant in the arctic, is the surety of change.
CARBON DIOXIDE & GREEN HOUSE WARMING
Up to this point, there has been no discussion of green house gasses and their effect on the arctic. Yet, one would expect the effect of the green house gas, carbon dioxide, to have it’s maximum effect in the deserts of the world. Water vapor is the most dominant green house gas except for locations where its concentration is low. Water vapor is almost non-existent in the atmosphere above the deserts of the world, so the second most prevalent green house gas, carbon dioxide, should predominate.
In addition, because of the very nature of green house warming, one would expect the effect to be most pronounced in the arctic and Antarctic. The reason for this is that the air temperatures at the lower elevations are increased most with carbon dioxide and the green house effect, and it is expected that this will melt the ice faster. The only data that could be found was from a projection of temperature rises due to carbon dioxide forcing. These were not actual data collected, but rather a projection. Temperatures in Alaska are rising faster than other locations in the USA, but the exact cause for that is not known with certainty.
The author of this article has shown in another article that the excess heating for the northern hemisphere can be explained by the way in which the two halves of the earth rotate about the sun. Once again, this is a theoretical treatment and not collected data.
OVERALL CONCLUSIONS
The overall conclusion to be derived from this presentation is that the ice coverage in the arctic is a very complex subject and there are a myriad of influencing factors involved. A secondary conclusion is that at the present time, the ice coverage in the arctic is decreasing, some years faster than others. This is because there is more heat inputted to the Arctic Ocean than there is heat being dumped to outer space. No other overall conclusions seem warranted, although the author has ventured some predictions based on other work.