This is William Crump's Typepad Profile.
Join Typepad and start following William Crump's activity
Join Now!
Already a member? Sign In
William Crump
Recent Activity
R Gates: While your estimates of total ice for September 2012 appear accurate, I am not able to confirm the statement that 80% of the September ice pack is MYI using the September ice age graph provided by the NSIDC at: http://nsidc.org/images/arcticseaicenews/20111004_Figure6.png While there was a dramatic decline in Arctic ice in 2007, the decline of first year ice was not as severe as you indicate. First and second year ice increased slightly as a percentage of the Arctic ice pack and made up approximately 60% of the ice pack in September of 2007! In September of 2008, there was a dramatic increase over 2007 for the portion of the ice pack that consists new ice that survived to become first year ice as the percentage of the ice pack that consisted of first year ice went from 40% to 60%. In September of 2011, more than 40% of the ice pack consisted of new ice, which became first year ice. First and second year ice combined made up more than 70% of the September 2011 ice pack. This was down slightly from 2009 and 2010 when this ice made up almost 80% of the September ice pack. Your comment leaves the impression that FYI is currently melting out 100% and is declining at a faster rate than MYI. This is not correct. Significant portions of newly formed ice are surviving the September melt to become first year ice. All indications are that MYI is declining faster than FYI. For example, see the study released earlier this year by Josephino Comiso titled "Large Decadal Decline of Multiyear Ice". http://www.nasa.gov/topics/earth/features/thick-melt.html "In the new study, Comiso and colleagues looked at multi-year ice that had made it through at least two summers. They wanted to see how it diminished with each passing winter over the past three decades. Results showed that the extent of multi-year ice, which includes areas of the Arctic Ocean where multi-year ice covers at least 15 percent of the water's surface, is shrinking at a rate of 15.1 percent per decade." The oldest and thickest portions of the ice pack show the greatest decline as ice older than three years appears to only make up about 10% of the ice pack in September of 2011. The September 2012 chart will be available in early October, but I would expect to see more than 40% of the ice pack to be newly formed first year ice and more than 20% of the ice pack will be second year ice.
Toggle Commented Aug 10, 2012 on Peeking through the clouds 3 at Arctic Sea Ice
R Gates: Where did you get the 80% MYI figure for September? The link below from Maslanik tells a much different story. Most of the ice in September consists of first and second year ice not MYI. http://nsidc.org/images/arcticseaicenews/20111004_Figure6.png
Toggle Commented Aug 10, 2012 on Peeking through the clouds 3 at Arctic Sea Ice
THICKER ICE IS DISAPPEARING FASTER THAN THINNER ICE NASA study reveals that the oldest and thickest Arctic sea ice is disappearing at a faster rate than the younger and thinner ice at the edges of the Arctic Ocean’s floating ice cap. http://www.nasa.gov/topics/earth/features/thick-melt.html Comiso found that perennial ice extent is shrinking at a rate of -12.2 percent per decade, while its area is declining at a rate of -13.5 percent per decade. These numbers indicate that the thickest ice, multiyear-ice, is declining faster than the other perennial ice that surrounds it. As perennial ice retreated in the last three decades, it opened up new areas of the Arctic Ocean that could then be covered by seasonal ice in the winter. A larger volume of younger ice meant that a larger portion of it made it through the summer and was available to form second-year ice. This is likely the reason why the perennial ice cover, which includes second year ice, is not declining as rapidly as the multiyear ice cover, Comiso said. "The rapid disappearance of older ice makes Arctic sea ice even more vulnerable to further decline in the summer, said Joey Comiso, senior scientist at NASA Goddard Space Flight Center, Greenbelt, Md., and author of the study, which was recently published in Journal of Climate."
Toggle Commented May 15, 2012 on PIOMAS May 2012 at Arctic Sea Ice
Pete Williamson: Recovery mechanisms of Arctic summer sea ice S. Tietsche,1 D. Notz,1 J. H. Jungclaus,1 and J. Marotzke1 http://www.seas.harvard.edu/climate/seminars/pdfs/Tietsche_GRL_2011.pdf is the source of the recovery period, which they say is two years. 4. Conclusions [23] In our perturbation experiments, we observe how different feedbacks in the Arctic compete to enhance or dampen a strong negative anomaly in sea ice, equivalent to a strong positive anomaly in oceanic heat content. In summer, the oceanic heat anomaly is enhanced by the ice–albedo feedback, but in winter the excess oceanic heat is lost to the atmosphere due to a lack of insulating sea!ice cover. This leads to an anomalously warm atmosphere, which in turn causes increased heat loss by longwave radiation at the top of the atmosphere and decreased heat gain by atmospheric advection from lower latitudes. A lasting impact of the ice–albedo feedback is not possible because the large scale heat fluxes quickly adapt to release the excess oceanic heat from the Arctic. [24] Hence, we find that even dramatic perturbations of summer sea ice cover in the Arctic are reversible on very short time scales of typically two years. This suggests that a so called tipping point, which would describe the sudden irreversible loss of Arctic summer sea ice during warming conditions, is unlikely to exist. [25] These results also have implications for the value of sea ice initial conditions for climate predictions on decadal time scales: if even the strong anomalies in sea ice cover that we examine here are reversible within a few years, then small errors in sea ice initial conditions should not affect the predictions significantly. Intrinsic memory of the thin Arctic sea ice cover of the 21st century seems to span only a few years.
Toggle Commented May 14, 2012 on PIOMAS May 2012 at Arctic Sea Ice
Here is the graph for thickness for 2004 through 2008. http://www.nasa.gov/images/content/365869main_earth2-20090707-full.jpg
Toggle Commented May 11, 2012 on PIOMAS May 2012 at Arctic Sea Ice
This graph shows the volume of first year ice was increasing in winter from 2004 to 2008. This is not the same as thickness as extent was likely expanding. http://www.nasa.gov/images/content/365871main_earth3-20090707-full.jpg
Toggle Commented May 11, 2012 on PIOMAS May 2012 at Arctic Sea Ice
Rob Decker: Is NAME a forecasting tool or is it a model for estimating current ice conditions? I thought it was the latter. Maslowski's "forecast" in Figure 9 appears to be an extrapolation - line drawing exercise and not a forecast based on future data points from a model of physical processes.
Toggle Commented May 11, 2012 on PIOMAS May 2012 at Arctic Sea Ice
While I am working on a graph, please see the following articles; http://www.agu.org/pubs/crossref/2007/2007GL032043.shtml http://www.agu.org/pubs/crossref/2011/2011GL047735.shtml
Toggle Commented May 11, 2012 on PIOMAS May 2012 at Arctic Sea Ice
Chris: I am struggling with the Bitz & Roe article, but I will get through it. This is the same Bitz from the comment letter on Real Climate about the perils of extrapolation. How does the late start to the melt season in 2010 affect the loss numbers, did this pile up extra thin young ice that was quickly removed from the system once the melt season started? If the transition from a mainly MY ice pack to a FY ice pack is nearly complete, does this mean the volume decline should flatten out? I understand your concern that freak weather conditions could result in an abnormally low single year, but the Bitz & Roe paper and the Tietsche paper appear to suggest that any such single year decline would not result in a switch to a permanently ice free Arctic at the minimum. Maslowski and other line drawers appear to be stating that the "ice free" status as they definite it will constitute a permanent feature once it occurs. These particular papers appear to suggest that single year perturbations do not result in a permanent "ice free" status and that a limited rebound can occur as subsequent year ice growth exceeds subsequent year ice melting and a new equilibrium point is obtained. Thanks for the posts. Perhaps what is needed is a standardized definition of "ice free" (I do not mind excluding the coastal areas of the Canadian Archipelago and Northern Greenland from areas to be considered in the determination of "ice free"). Included in this definition should be a requirement that this "ice free" state should occur in at least three consecutive years.
Toggle Commented May 10, 2012 on PIOMAS May 2012 at Arctic Sea Ice
FrankD: It is not just my flimsy opinion, it is the opinion of the scientists who create the volume data being used in the various line drawing exercises on this web site. They have posted a warning that their data should not be used in the fashion displayed on this website and have directly referenced the PIOMAS Monthly Arctic Ice Volume with exponential trend graph as not being a valid forecasting method. You, of course, may choose to ignore the warning of these scientists and go your own way, but you risk becoming as irrelevant as Steve Goddard's web site, which would be most unfortunate. I am submitting as proof that my position is not the mere ravings of a mad man (although I may fit part of that description) the comment letter posted on Real Climate at: http://www.realclimate.org/index.php/archives/2012/04/arctic-sea-ice-volume-piomas-prediction-and-the-perils-of-extrapolation/ The authors of this letter say that extrapolations made in the manner that the various line graphs displayed on this web site using the PIOMAS volume data are not valid. They lay out in better detail than I why such lines are not valid. Please explain why these scientists are wrong and you are correct. Since these people are the scientists involved in the PIOMAS volume data, I will accept their opinion over the various bloggers on this site, although I appreciate the information and points often brought to light on this site. They warn that: "Natural variability at these time scales (order of 30 years) may very well make prediction by extrapolation hopeless." In summary they state: "In summary, we think that expressing concern about the future of the Arctic by highlighting only the earliest estimates of an ice-free Arctic is misdirected. Instead, serious effort should be devoted to making detailed seasonal-to-interannual (initial-value) predictions with careful evaluations of their skill and better estimates of the climate-forced projections and their uncertainties, both of which are of considerable value to society. Some effort should also target the formulation of applicable and answerable questions that can help focus modeling efforts. We believe that substantially skillful prediction can only be achieved with models, and therefore effort should be given to improving predictive modeling activities. The best role of observations in prediction is to improve, test, and initialize models. But when will the Arctic be ice free then? The answer will have to come from fully coupled climate models. Only they can account for the non-linear behavior of the trajectory of the sea ice evolution and put longer term changes in the context of expected natural variability. The sea ice simulations in the CMIP5 models are currently being analyzed. This analysis will reveal new insights about model biases, their causes, and about the role of natural variability in long-term change.It is possible that this analysis will change the predicted timing of the “ice free summers” but large uncertainties will likely remain. Until then, we believe, we need to let science run its course and let previous model-based predictions of somewhere between “2040 and 2100″ stand” The authors of the letter are: Axel Schweiger "Dr. Schweiger is the current chair of PSC. His research focuses on the interaction of sea ice with clouds and radiation. He is using satelite data, models and in-situ observations to improve our understanding of sea ice and cloud variability. He has developed the PSC Arctic Ice Volume Page . He has been working to impove estimates of the surface radiation balance in the Arctic. To this end he has been developing and evaluating satellite-based algorithms. He has been assembling the TOVS Polar Pathfinder data set, a 20-year data set of polar temperature, humidity profiles and cloud information. Previous research includes work on microwave-based sea ice concentration algorithms and the application of artificial intelligence methods to remote sensing problems. Dr. Schweiger has been with the Polar Science Center since 1992." http://psc.apl.washington.edu/wordpress/people/investigators/axel-schweiger/ Ronald W. Lindsay, Senior Principal Physicist for the University of Washington APL http://www.apl.washington.edu/people/profile.php?last=Lindsay&first=Ron Cecilia Bitz I am an Associate Professor in the Atmospheric Sciences Department, an Affiliate Physicist for the Polar Science Center, and part of the Program on Climate Change, all at University of Washington. http://www.atmos.washington.edu/~bitz/ You may choose to follow the line drawn in Figure 9 of the Maslowski article using October-November mean ice volume which indicates that the mean for these months will be zero (or close to it) by 2016 (or 2019 if you need some extra years). I find this claim to be unrealistic when the satellite PIOMAS data show increasing volume of 8,000 km3 for the last two years for the period September 1 to November 30 as the ice begins its annual rebound from a low of 4,000 km3 to 12,000 km3 at November 30. No force discussed on this web site is going to reverse this gain of 8,000 km3 over this period.
Toggle Commented May 10, 2012 on PIOMAS May 2012 at Arctic Sea Ice
Kevin: If by "privileged" you mean that the northernmost region of the planet has lower air temperatures and water temperatures than regions in lower latitudes and therefore less ice melting and more ice formation occur in the Arctic Basin than in other regions, then certainly ice in the Arctic basin is "privileged". Also, this region receives a significant influx of ice from the Laptev Sea, close to 500,000 km2 per year. While the Greenland Sea also receives a large influx of ice through the Fram Strait, this ice flow is on a "death march" as less than 100,000 km2 of ice survives in the Greenland Sea at the minimum (I am not certain any of this transported ice survives the minimum as it looks like the ice at the minimum in the Greenland sea is ice that is close to the shoreline and the transported ice, which is further off the coast just disappears in the fascinating video mentioned in Bob Wallace's comment). http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/recent365.anom.region.5.html By contrast, a higher percentage of ice transported into the Arctic Basin from the Laptev Sea survives at the minimum. This is another factor that maintains the Arctic Basin above 2.5 million km2 at the minimum since 2007. If the Laptev Sea ice fails to form, then there will be steep declines in the Arctic Basin. http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/recent365.anom.region.8.html What is the area trend line for the Arctic Basin region (which has an area in excess of 2.5 million km2 at the minimum) per the following graph from Cryosphere today: http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/recent365.anom.region.1.html When does this trend line suggest that the Arctic basin will become "ice free" - say below 500,000 km2? And yes, if we had thickness data for the various types of ice in the Arctic Basin I would prefer trend lines based on these measures rather than area, but this data is not available so I will make do with the data at hand.
Toggle Commented May 10, 2012 on PIOMAS May 2012 at Arctic Sea Ice
Peter Ellis: The current state of the Arctic ice appears to be doing exactly what you state in items a. and b. If you extrapolate the trends in the ice age charts below, what is the trend that you see? Please note that when I use the term multi-year ice I am referring to ice that is more than 2 years old. Perhaps some of the line drawers can provide a chart for ice in the various age categories that incorporates the percentage of ice ages in the chart and the decline in area/extent for each year. This would at least provide a clearer picture of the area trend line for ice younger than 2 years and ice older ice. Both points a. and point b. appear to be confirmed by the NSIDC charts and narrative's in various press releases. I agree this does not make intuitive sense that younger ice is able to regrow and regenerate and survive better than multi-year ice; particularly the subset of multi-year ice that is made up of ice that is older than three years, but isn't that exactly what the charts below and the NSIDC narrative are saying? Per the NSIDC, the age of ice in September of 2011 and March of 2012 are close to the situation where the Arctic is dominated by thinner young ice. The March 2012 ice age chart shows that ice 2 years old and under makes up 80% of the ice and the NSIDC says that ice more than 4 years old has gone from 25% of the ice pack to 2% of the ice pack. Per NSIDC: "However the oldest, thickest ice, more than four years old, continued to decline. Ice older than four years used to make up about a quarter of the winter sea ice cover, but now constitutes only 2%." March 2012 chart is at: http://nsidc.org/arcticseaicenews/files/2012/04/Figure5.png The September 2011 Ice age chart is at: http://nsidc.org/images/arcticseaicenews/20111004_Figure6.png On the September 2011 chart, ice that is 4 years old or older has gone from close to 50% of the ice pack at the minimum to less than 10% of the ice pack. The NSIDC noted in their October 2011 press release the following: "First- and second-year ice made up 80% of the ice cover in the Arctic Basin in March 2011, compared to 55% on average from 1980 to 2000. Over the past few summers, more first-year ice has survived than in 2007, replenishing the younger multi-year ice categories (2- to 3-year-old ice). This multi-year ice appears to have played a key role in preserving the tongue of ice extending from near the North Pole toward the East Siberian Sea. However, the oldest, thickest ice (five or more years old) has continued to decline, particularly in the Beaufort and Chukchi Seas. Continued loss of the oldest, thickest ice has prevented any significant recovery of the summer minimum extent. In essence, what was once a refuge for older ice has become a graveyard. " Given that the older thicker multi-year ice has declined more than the thinner younger ice, isn't it possible that the volume decline lines are more indicative of when the Arctic will become "ice free" of the oldest thickest ice (with the exception of the older ice that winds drive up against the Canadian Archipelago and Northern Greenland) and not indicative of when the Arctic will be "ice free" of ice less than 3 years old? The following is my guess for how the September 2012 age of ice chart will play out: 5 plus years old will continue to decline as it is not being replenished as fast as it is being melted and transported out of the Arctic. This type of ice will continue to be a marginal portion of the ice - under 5% of the September ice pack 4 year old ice will increase in percentage because at the September 2011 minimum there was more three year old ice in 2011 than there was in 2010, but it will not be a significant proportion of the ice pack (I am guessing it will be under 5% - yes I know this is a simplistic assessment 3 year old ice will remain close to the 2011 percentage since the amount of 2 year old ice at 2011 is about the same as 2010. I would guess that it may decline slightly as the Arctic no longer appears to favor making older ice. 2 year old ice will decline in percentage because there was less one year old ice at September 2011 than 2010. Less first year ice in the prior year means less second year ice the following year. Ice making it to its first September, one year old ice, will increase in percentage; but this increase is not occurring because the Arctic is recovering, it is simply a function that there is less area taken up by older ice so there is more area for first year ice to develop. As for volume and area/extent, the weather will determine what these values are. If we get weather like 2008 and 2009, whatever that is, we may see a slight increase in the volume at the minimum. If we get weather like 2007 and 2010 then there will be a drop in volume, but not as much as in the past because there is less of the thickest ice to be blown out of the Arctic down the Fram Strait.
Toggle Commented May 10, 2012 on PIOMAS May 2012 at Arctic Sea Ice
Daniel: Yes it is winter indeed. I think you are getting at the problem with the graph in Figure 9, it flies in the face of the physical process of winter. I am saying winter will have a stronger influence on the October-November mean volume than the lines drawn in Figure 9 for the October-November mean which show no volume by 2016 or 2019 at the latest. What do you believe will win out in determining the future October-November mean volume, winter or the lines in Figure 9? As for Wipneus diagram, it is very pretty, but it does not factor in that prior volume losses have occurred due to a type of ice that is virtually non-existent in the Arctic and therefore it is incorrect to assume that the rate will continue. You have yet to show me the physical processes that will cause the ice in the region Cryosphere Today calls the Arctic Basin to suddenly drop from the 2.5 million km2 level it has maintained at the minimum since 2007 to almost nothing in the next three years. Other than saying the lines reflect historical rates, none of you have shown what physical processes will occur in future years that will allow this rate of decline to be sustained. Even though the models have not matched the actual rate of decline, it does not necessarily follow that they do not have the correct rate of decline for future periods. You can draw all the lines you like, but they prove nothing unless they are backed by a model employing the physical processes that justify the shape of the lines; otherwise, they are just lines.
Toggle Commented May 9, 2012 on PIOMAS May 2012 at Arctic Sea Ice
Chris: I agree that I do not have a mechanism for why the oldest ice is disappearing at such a high rate and Bob Wallace is correct that the ice is moving from other regions into the Arctic Basin. The diminished Barents Sea should not impact this analysis as winds generally do not push this ice into the Arctic Basin. In this regard, perhaps the Laptev Sea is worth watching since: "The Laptev Sea is a major source of arctic sea ice. With an average outflow of 483,000 km2 per year over the period 1979–1995, it contributes more sea ice than the Barents Sea, Kara Sea, East Siberian Sea and Chukchi Sea combined. Over this period, the annual outflow fluctuated between 251,000 km2 in 1984–85 and 732,000 km2 in 1988–89. The sea exports substantial amounts of sea ice in all months but July, August and September.[7]" (sorry this is from wikipedia). I am just repeating the observation by the NSIDC that the oldest and thickest type of multi-year ice has almost disappeared, regardless of the mechanism that caused it to disappear: "Ice older than four years used to make up about a quarter of the winter sea ice cover, but now constitutes only 2%." http://nsidc.org/arcticseaicenews/files/2012/04/Figure5.png This should have a disproportionate effect on the volume loss and thickness figures making the use of average figures for the Arctic as a whole or extrapolations based on continuing the historical rate of volume loss suspect. The contributions that the loss of this thickest and oldest ice made to the PIOMAS volume decline from 2007 to 2011 can not be repeated because this type of ice is no longer there to be melted or otherwise transported out of the Arctic. It will take several years of PIOMAS data and not just one more month to see if there is a slow down in the rate of volume loss. That the ice will diminish is not in doubt, we are only discussing how fast it will go.
Toggle Commented May 9, 2012 on PIOMAS May 2012 at Arctic Sea Ice
Chris and Rob: In addition to the observation that it is easier to create new ice where none exists at the start of the freeze-up season and to add thickness to younger ice than it is too add thickness to old ice, I was wondering if you can quantify how much of the disparate volume loss in 2007 and 2010 can be explained by the advection of thick multi-year ice out of the Arctic through the Fram Strait rather than just melting in place factors? Surface melt should be relatively the same regardless of whether the ice is thick or thin (please let me know if this is an incorrect assumption); therefore, the difference in the amount of decline in thickness between different types of ice must be coming from bottom melting. Since the warmer arctic waters are too deep at 100 meters to be in direct contact with Arctic ice the difference in bottom melt must be coming from a different mechanism. Clearly it takes more heat input to melt more of the thicker ice than is being applied to the reduced melting of thin ice. So what are the possible mechanisms that would allow greater exposure to heat for thick multi-year ice compared to thinner new and first year ice? Does the thicker ice present a more irregular shape and greater profile that presents greater surface areas for bottom melting than the more uniform depth of thin ice? Are the mechanics of maintaining a thin insulating layer of cold water better for thinner ice than it is for thicker ice which has irregular depths? Are there other possibilities for why greater heat is reaching multi-year ice than new and first year ice?
Toggle Commented May 9, 2012 on PIOMAS April 2012 at Arctic Sea Ice
Chris: I see the issue you are working on and I have wrestled with it myself. "In the above figure you can ignore ExtentThickness as I’ve moved over to calculating AreaThickness, using area alone. Kevin O’Neill (a commenter at my blog) and I have been referring to the relatively flat area through the Summer as the ‘roof’. Its implication being that during the Summer volume loss has been largely accounted for by loss of area, hence thickness changes little. However in 2010 and 2011 this changed, with loss of thickness being ‘needed’ to account for the loss of volume. As the above graph shows this is unusual behaviour in the context of the full PIOMAS series. Using NCEP/NCAR reanalysis the only weather that stands out as being anomalous is a high pressure system over the Arctic in 2010 presumably related in some way to high temperatures. However as the area was ice covered at that time increased insolation wouldn’t have caused high temperatures, although inflow of air or large leads/polnyas could. Spring 2011 was rather unremarkable from what I can see." Please consider that a possible explanation for what you are seeing is the disparity in the types of ice that are declining and how their decline is contributing to the average calculation. An underlying assumption of the "thickness" analysis is that it assumes that all ice types are declining in volume and thickness at uniform rates. I do not think this is the case. The disproportionate losses of thick multi-year ice, which the NSIDC indicates below have nearly disappered: http://nsidc.org/arcticseaicenews/files/2012/04/Figure5.png "After the near-record melt last summer, second-year ice declined again, but some of the ice that had survived the previous few summers made it through another year, increasing the proportion of third- and fourth-year ice. However the oldest, thickest ice, more than four years old, continued to decline. Ice older than four years used to make up about a quarter of the winter sea ice cover, but now constitutes only 2%. First-year ice (0 to 1 years old) this year makes up 75% of the total ice cover, the third highest at this time of year in the satellite record. In 2008 the proportion of first-year ice was 79%, and in 2009 it was 76%." may be driving a disproportionate portion of the volume and thickness decline. Very thick multi year ice has been replaced by thinner first and second year ice and the thickness of the remaining multi-year ice is considerably thinner on average than the thickness of multi-year ice that used to cover the Arctic. If a disproportionate amount of the thickness decline is attributable to multi-year ice, then the thickness of first and second year ice is not declining as fast as the graph would indicate. What is needed is a direct measurement of the rate of decline in the thickness of new (ice formed in the current year) and first year ice (ice that survived the prior September). If the thickness of this ice is not declining as fast as the rate of decline from the average of all types of ice, then it may be possible to maintain a much larger area of thin young ice at the minimum for a much longer period of time than the extrapolation of Artic wide volume and average thickness data would indicate. The last area to disappear will likely be the Arctic Basin region as defined by Cryosphere today. http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/recent365.anom.region.1.html A substantial majority of this ice appears to be first and second year ice. The big unknown is how fast the thickness of this young ice that survives the September minimum is diminishing. If the thickness of new and first year ice in the Arctic Basin is declining at a slower rate than the Arctic as a whole, then that would explain why the area of the Arctic Basin ice has not declined since 2007 in the face of the extreme volume decline experienced by the Arctic as a whole. Is there any manner in which your area thickness chart can be altered to generate a thickness decline curve for young ice in the Arctic Basin?
Toggle Commented May 9, 2012 on PIOMAS May 2012 at Arctic Sea Ice
Daniel: NAME is a model to estimate current volume conditions not a forecasting tool. The data points in Figure 9 for NAME stop in 2005. There are no data points from NAME in Figure 9 after 2005 - notice how the blue line and dark blue asterisks stop in 2005. Maslowski uses Kwok current estimates of volume (light blue asterisks and magenta asterisks) for data points in 2004 through 2008. There are no data points after 2008. You should give the article a good reading. I too thought Maslowski had a model that generated multiple scenarios for forecasting future ice conditions until I read the article provided by Kevin and realized that this characterization is a sham. NAME is being used to generate historical data points from which a linear extrapolation is performed by Maslowski. The NAME model does not generate future data points in the manner that the models discussed in the Maslowski article create multiple forecasts of future ice conditions. Maslowski's article does a good job of slamming all predictive models; however, Maslowski is just line drawing based on historical NAME and Kwok data points. His 2016 claim is not based on a model simulation of 2016 conditions, it is based on a cherry pick of data points and a linear extrapolation with no consideration of the complex and dynamic factors affecting ice behavior. The models discussed in the article generate multiple future outcomes. Maslowski is just drawing lines for his "prediction.".
Toggle Commented May 9, 2012 on PIOMAS May 2012 at Arctic Sea Ice
Chris: I am taking my time to go through the thoughtful information you have provided. With natural variability, I guess it is possible for the ice to drop though a 1,000,000 km2 threshold (a size 43.6% bigger than Texas, which seems rather large to me as my state, Maryland, is just over 32,000 km2 including the water) by 2020, and it would only take a volume of 500km3 if the ice is half a meter thick to generate an area/extent this large, but it does not appear that the Arctic would maintain this condition year in and year out, so I would not say the Arctic reached a steady state of ice free existence. What I am looking for is a data set on volume or better yet, thickness, for the region Cryosphere today calls the Arctic Basin. http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/recent365.anom.region.1.html I would like to see a trend line for the thickness of first year ice for this region that shows that the extent will fall below 500,000 km2 (an area larger than California) before joining the "ice free" by 2016 to 2019 bandwagon. I can not find such a data set, but the area figures from Cryosphere today indicate that the region is maintaining a minimum area of approximately 2.5 million km2 since the big plunges in 2006 and 2007. I do not think the Arctic wide volume data set is appropriate for generating a prediction of when the Arctic Basin will be ice free as it includes many regions which have already reached a zero volume level at the minimum and it is skewed by the disproportionate loss of thick multi-year ice. The Arctic Basin will not be "ice free" until it reaches a point that the ice that forms in winter does not survive the melt season. Provide a graph of first year ice thickness for the Arctic Basin that shows that first year ice that forms in winter does not reach a sufficient thickness to allow it to consistently survive the melt season by 2019 and I can agree with the Maslowski extrapolation.
Toggle Commented May 9, 2012 on PIOMAS May 2012 at Arctic Sea Ice
Daniel: What Maslowski model are you talking about? There is no Maslowski predictive model in the article posted by Kevin for future ice prediction. Maslowski took data points for mean October-November ice volume per a model he uses to estimate current ice volume for the period 1996 through 2007 and drew a straight line. The NAME data points are estimates of actual ice volume, not projections of future ice conditions. He added data from Kwok et al for the period 2004 to 2008 to fill in missing data from NAME. Note the Kwok ice volume estimates for 2004 and 2005 are higher than the NAME volume estimates Maslowski calculated a decline of 1,120 km3 per year using the period 1996 through 2007 (he left out 1979 through 1996 since it showed an increasing volume amount per the red dashed line in Figure 9. He also states the standard deviation is plus or minus 2,353 km3 which is an extremely large range), and stated that the 2007 October-November mean was under 9,000 km3. At this rate of decline, he stated that it would reach a zero point in nine years (2016). There is no model. Please answer the question, the data points used in Maslowski's line drawing exercise used mean estimated values for October-November ice volume. His line predicts the arctic will be "ice free" in the months of October and November by 2016 plus or minus 3 years. Since the ice volume has expanded in every month following August in every year of the satellite record, I do not think a claim of an ice free October and November by 2019 is supportable.
Toggle Commented May 9, 2012 on PIOMAS May 2012 at Arctic Sea Ice
Phil: Amazing graphic, but I am not sure it answers my question as it only shows 2012 flow. Below is 2011 flow: http://www7320.nrlssc.navy.mil/hycomARC/navo/arcticictn_nowcast_anim365d.gif It looks to my biased and untrained eye that the thin stuff is moving fast while the thicker ice along the coast is not moving that much. Also, there does not appear to be much of the thicker stuff. I am relying on NSIDC statement that much of the oldest and thickest ice is gone. http://nsidc.org/arcticseaicenews/2012/04/ Is there a 365 day graphic for 2007 and 2010 that I can compare with 2012?
Toggle Commented May 9, 2012 on PIOMAS May 2012 at Arctic Sea Ice
Wipneus: The PIOMAS graph is showing that 2012 has been at the same level as 2011 or a little above it. Why is the 2012 projection of September volume so much lower than 2011? Are the big drops in 2007 and 2010 skewing the line? The second largest manner in which the Arctic loses ice volume is through the Fram Strait. The year to year loss through the Fram Strait is highly variable. See http://www.cgd.ucar.edu/cas/cdeser/Docs/climdyn_tsukernik-framstrait.pdf (Vinje 2001; Brummer et al. 2001, 2003; Kwok 2009). Given that the ice is thinner in 2012 than prior years and the diminished ice cover in the Barents Sea, is it possible that less ice volume will be lost through the Fram Strait in 2012 than in previous years and this will operate as a negative feedback that prevents ice volume from falling as quickly as it has in the past?
Toggle Commented May 8, 2012 on PIOMAS May 2012 at Arctic Sea Ice
Daniel: Do you believe the extrapolation in Figure 9 that the Arctic will be ice free in October-November of 2016? That is what it shows as the "ON" at the top of the graph means October-November. Kevin: Thanks for providing this article. Maslowski does a good job of showing that the existing models have understated observed ice loss, and describes the difficulties of modeling a system as complex as Arctic ice. How he makes the leap of faith that figure 9 is a better method for prognostication of future Arctic ice conditions rather than use improved models, I do not know. His "prediction" is a simple linear extrapolation based on a limited data set (one might say cherry picked) of October-November mean volume for 1996 through 2007 which he states shows a decline of 1,120 per year. Maslowski excludes data for 1979 to 1996 which showed an increase in volume - the dashed red line in figure 9. Starting with a 2007 October-November mean of under 9,000 km3 he derives that the Arctic will be ice free in 9 years from 2007. The standard deviation amount for this claim is plus or minus 2,353 km3 per year. I think Wipneus has a better idea than Maslowski, but I doubt we will get a model that can duplicate the chaotic system of natural variability entailed in the steep decline up to 2011. The volume declines of 2007 and 2010 were steep, but we also saw in 2008 and 2009 that volume increases are possible over the short term. Current ice conditions may not matter in making a long term prediction as this cautionary statement from the Real Climate article illuminates. "The seasonal prediction issue and the prediction of the long-term trajectory are fundamentally different problems. Seasonal prediction, say predicting September ice extent in March, is what is called an initial value problem and the September ice extent depends both on the weather, which is mostly unpredictable beyond 10 days or so, and the state of the ocean and sea ice in March. Improving observations to better characterize that state, and improving models to carry this information forward in time is our best hope to improve seasonal predictability. The prediction of the long-term trajectory, depends on the climate forcing (greenhouse gases, aerosols, solar variability) and how the model responds to those forcings via feedbacks. A recent model study showed that the crossover between initial-value and climate-forced predictability for sea ice occurs at about 3 years (Blanchard-Wrigglesworth et al. 2011). In other words, a model forgets the initial sea ice state after a few years at which point the main driver of any predictable signal is the climate forcing. In fact, coupled model simulations have shown that even removing all the sea ice in a particular July has little lasting impact on the trajectory of the ice after a few years (Tietsche et al. 2011)." For now I will stick with a later date for an ice free Arctic at the minimum than 2016 and look for a better data set upon which to make an armchair prediction. Until then, I will continue to defer to the likes of Wang and Overland who forecast an ice free Arctic by 2037 in a 2009 research paper.
Toggle Commented May 8, 2012 on PIOMAS May 2012 at Arctic Sea Ice
Wipneus: Thanks for the error band graph and the Hamilton quote. Both are quite informative. The published scientific literature on arctic ice does not appear to support an "ice free" Arctic within the time frame of the chart you referenced in your comment. Published projections, though with varying definitions of what constitutes "ice-free", all project an ice-free Arctic ocean somewhere between 2037 (Wang and Overland, 2009) and the end of the century. Given the amount of natural variability, an accurate prediction of the minimum volume level in September of 2012 is more a matter of luck than science. I do not see why a prediction based on an extrapolation using an exponential curve fit and historical volume information from the PIOMAS model is a valid method of predicting 2012 volume levels or when "ice free" conditions will occur. To the best of my knowledge, there is no scientific study based on the physical forces and processes that are expected to affect future Arctic ice conditions which supports the use of an exponential line as the best fit for predicting future changes in Arctic ice volume. The models appear to show a future rate of ice decline that is many orders of magnitude slower than the exponential line used in the chart referenced in your comment. This slower rate of decline is due to the increasing influence of negative feedback factors, such as loss of ocean heat due to loss of the insulating impact of Arctic ice. Even if the exponential fit in the error band chart is appropriate for predicting the 2012 minimum volume, it is not certain that 2012 will have a lower volume amount than 2011. The error band chart appears to support the a possibility of higher volume in 2012 than 2011. The line drawn on the chart above makes it appear that a 25% decline in ice volume is almost certain to occur in 2012. While this is one possible outcome, it is not the only outcome that lies within the 95% probability range. The error band chart appears to say that there is a 95% confidence level that the 2012 minimum volume will be between 5,000 km3 - which is an increase of 1,000 km3 from 2011 - and 500 km3 - a virtually "ice-free" state, a range of 4,500 km3. The wide band of possible outcomes makes the use of the exponential line for projecting the 2012 minimum volume nearly worthless. Rather than draw a single line, the 2012 prediction line should start from current ice conditions at a particular date and show an expanding cone of possible outcomes for 2012 volume that would be within the 95% probability range. If this is a correct analysis of the error band chart, then to draw only a single line as a statistical projection of the possible level for 2012 is very misleading. If I am misreading the error band chart, please let me know.
Toggle Commented May 7, 2012 on PIOMAS May 2012 at Arctic Sea Ice
Wipneus: If the "statistical error bars are quite large" then how meaningful is the 2012 projection line? Instead of a single line, the chart should show the whole band of possibilities represented by the error bars. In looking at "actual" amounts from the PIOMAS model, 2012 appears to be closely following 2011 as noted by idunno. What physical processes in terms of winds and temperature and ice transport would have to be different in 2012 from 2011 in order for the September 2012 minimum to be 75% (based on a drop from 4,000 km3 to 3,000 km3)of the volume at the minimum in 2011? Any one year drop of the magnitude suggested by the 2012 projection line in the September minimum volume will have more to do with natural variability than long term CO2 forcing. Axel Schweiger, Ron Lindsay, and Cecilia Bitz have posted a comment letter on Real Climate "Arctic Sea Ice Volume: PIOMAS, Prediction, and the Perils of Extrapolation" that sets forth why such line drawing is suspect when dealing with a complex non-linear system like arctic ice. Two excerpts from the comment letter follow: "So does it make sense to extrapolate sea ice volume for prediction? In order to do a successful extrapolation several conditions need to be met. First, an appropriate function for the extrapolation should be chosen. This function needs to either be based on the underlying physics of the system or needs to be justified as appropriate for future projections beyond just fitting the historical data. But what function should one choose? Since we don’t really have data on how the trajectory of the Arctic sea ice evolves under increased greenhouse forcing, model projections may provide a guide about the shape of appropriate function. Clearly, linear, quadratic or exponential functions do not properly reflect the flattening of the trajectory in the next few decades seen for example in the CCSM4 (Fig 3). The characteristic flattening of this trajectory, at first order, arises from the fact that there is an increasingly negative (damping) feedback as the sea ice thins described by Bitz and Roe (2004) and Armour et al. (2011). The thick ice along the northern coast of Greenland is unusually persistent because there are on-shore winds that cause the ice to drift and pile-up there. So extrapolations by fitting a function that resembles a sigmoid-shaped trajectory may make more sense, but even that, as shown in the figure, yields a much earlier prediction of an ice-free Arctic than can be expected from the CCSM4 ensemble." On the discussion of when the arctic will be "ice free" they caution: "But when will the Arctic be ice free then? The answer will have to come from fully coupled climate models. Only they can account for the non-linear behavior of the trajectory of the sea ice evolution and put longer term changes in the context of expected natural variability. The sea ice simulations in the CMIP5 models are currently being analyzed. This analysis will reveal new insights about model biases, their causes, and about the role of natural variability in long-term change.It is possible that this analysis will change the predicted timing of the “ice free summers” but large uncertainties will likely remain. Until then, we believe, we need to let science run its course and let previous model-based predictions of somewhere between “2040 and 2100″ stand” See the full comment at: http://www.realclimate.org/index.php/archives/2012/04/arctic-sea-ice-volume-piomas-prediction-and-the-perils-of-extrapolation/
Toggle Commented May 7, 2012 on PIOMAS May 2012 at Arctic Sea Ice
For the record, when I use the phrase "Arctic Basin" I am referring to the light green area on Cryosphere Today which is represented by this graph: http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/recent365.anom.region.1.html Other sources appear to call this the central Arctic Basin but the areas do not always appear to match up with each other and I find that confusing so I generally reference the discussion back to Cryosphere Today as a source. Thank you Daniel Bailey for correctly pointing out that certain areas where winds pile up ice will be the last to melt out, such as the sheltered northern lees of Greenland and the Canadian Archipelago. I have not focused on these areas as I have seen multiple sources As a week defense to my focus on the Arctic Basin, in most discussions on the topic of "ice free" the term has generally been described in a fairly loose fashion as being ice free except for the areas mentioned by Daniel Bailey or as being a state at the minimum when total ice extent/area falls below some threshold such as 300,000 km2 rather than the absolute zero ice situation of 0.00km2. The 8% figure is based on the information provided by NSIDC when they stated: "Ice older than four years used to make up about a quarter of the winter sea ice cover, but now constitutes only 2%." The 8% is derived by dividing 2% by 25%. The 92% figure for missing old ice is calculated by dividing 23% by 25%. I do not have the source of the 10 meters to 3.5 meters decline, but these appear to be accepted numbers based on information provided by other commentators on this excellent site.
Toggle Commented Apr 19, 2012 on Long-term regional graphs at Arctic Sea Ice