This is Rob Dekker's Typepad Profile.
Join Typepad and start following Rob Dekker's activity
Join Now!
Already a member? Sign In
Rob Dekker
Recent Activity
According to Wipneus' concentration maps : the 'course' (25 km^2) NSIDC concentration dropped very quickly to normal levels, but the high resolution (3.125 km^2) concentration is still anomalously high. What I make of that (Wipneus please correct me if I'm wrong) is that there is a lot of thin ice at the edges turning to slush rather quickly, but the interior of the pack is still quite "white" and resilient. I think the jury is still out there on where 2015's minimum will end up.
Toggle Commented Jun 16, 2015 on Melt Pond May 2015 at Arctic Sea Ice
I posted this elsewhere but I think it is more applicable here : Larry, thank you for posting these public belief findings. I think your statements to be spot-on, at multiple levels : On this and other factual questions, it seems likely that many people chose answers derived from their more general beliefs On this and other factual questions, it seems likely that many people chose answers derived from their more general beliefs I find evidence in that also on other subjects, such as MH17, Keystone XL, evolution theory and AGW. On a positive note, mis-beliefs often just go away once evidence is overwhelming. In that regard, I find it encouraging that your survey shows, that the opinion of "climate is changing due to natural causes" is for the first time dipping below 50%.
Thank you Neven, for a well thought out, balanced overview of the situation in the Arctic at this point. I appreciate David Schröder's team melting pond assessment this year There is no doubt that melting ponds greatly affect the melting season locally, and thus his data is important and very well appreciated. Did you obtain any information on how David Schröder's team determines this info ?
Toggle Commented Jun 15, 2015 on Melt Pond May 2015 at Arctic Sea Ice
Larry, thank you for posting these public belief findings. I find your statements to be spot-on, at multiple levels : On this and other factual questions, it seems likely that many people chose answers derived from their more general beliefs I find evidence in that also on other subjects, such as MH17, Keystone XL, evolution theory and AGW. On a positive note, mis-beliefs often just go away once evidence is overwhelming. In that regard, I find it encouraging that your survey shows, that the opinion of "climate is changing due to natural causes" is for the first time dipping below 50%.
navegante, yes, that is correct, especially since BOTH extent and area are running at record lows already.
Sorry to be late to the party (some trouble with NSIDC area numbers, and some uncertainty on snow numbers), but I just submitted my projection to ARCUS : 4.9 M km^2 with SD 470 k km^2. This estimate is based on linear regression of how 'dark' the Northern hemisphere was during April and May, as estimated by two variables : (1) Rutgers snow cover numbers from April and May and (2) NSIDC May numbers of (extent-minus-area) to estimate dark (leads and melting ponds. If I do the same linear regression excluding April snow numbers, I end up with a projection of 4.5, and a standard deviation of 520 k km^2, and honestly speaking, I think the April snow numbers just give a better SD because Northern Hemisphere snow numbers (including April) are following the normal linear down trend that can be expected from a globally warming planet. So I think my 4.9 is conservative, and I would not be surprised if I need to down-adjust that number steeply next month. Also note that even a simple linear trend obtains a SD of about 550 k km^2 or so, and thus I take my projection with a VERY LARGE grain of salt. June projection should be much better.
Bill Fothergill said It looks as though someone at NSIDC has done a bit of cut&paste without checking if the paragraph still makes sense. ... I would hazard a guess that, from Jan 2008 onwards, the offset value reduces to 0.029 - but it could certainly have been phrased better. Does that help? Yes, it does, and thank you for your reply. I also noticed the difference between the notes in the old and the new files, and that the new 'area' numbers are adjusted upward (by indeed about 0.3) for 2008 onward. So I think you are right that the new discontinuity is in 2008 only, and not in 2013. But I will contact the NSIDC helpdesk just to make sure. Wipneus said : Think of it: even if area equals extent every day of the month - 100% concentration within the ice extent, never any melt ponds or leads- grid cells that are only covered with ice during part of the month lead to an average concentration less than 100%. I had to read that sentence a couple of times, but I understand what you mean. And yes, this may be a problem for even my simple model, and it may explain differences between the NSIDC's area numbers and other data sets. Let me think about this a bit and then I will try to come up with a better method for determining "extent minus area" as a metric to determine leads and melting ponds. Thanks !
Also, I'd like some clear mind advice on something. I developed a super-simple model that uses Northern Hemisphere spring snow cover and a metric for melting ponds as a predictor for the September ice minimum : This model works very nicely, as it obtains better correlation numbers than plain extrapolation of the long term trend. For Northern hemisphere snow cover I use Rutgers Snow Lab monthly numbers, and for "melting ponds" (and other water close to ice) I use NSIDC numbers for (ice extent - minus - ice area) as a metric. This year, as Neven reports, ice "extent" in May is running very low, while ice "area" is sort of average. However, and here is the issue : NSIDC does not seem to share that observation. Here are the numbers : .. 2012 5 Goddard N 13.11 10.99 --> E-A = 2.12 Capie:A/E= 0.838 2013 5 Goddard N 13.08 11.20 --> E-A = 1.88 Capie A/E = 0.856 2014 5 Goddard N 12.77 10.99 --> E-A = 1.78 Capie A/E = 0.860 2015 5 NRTSI-G N 12.65 10.78 --> E-A = 1.87 Capie A/E = 0.852 Numbers obtained from here : So it seems that NSIDC does not really rate May 2015 as specifically high on Capie index, nor out of the ordinary for the "extent minus area" indicator that I use as a metric for "melting ponds" in May. So, for starters, I wonder why that difference in data sets (between NSIDC and IJIS extent and CT area) came about. Second, the problem may be related to the way that NSIDC calculates their sea ice "area" numbers. From the same NSIDC txt file I linked above : The "extent" column includes the area near the pole not imaged by the sensor. It is assumed to be entirely ice covered with at least 15% concentration. However, the "area" column excludes the area not imaged by the sensor. This area is 1.19 million square kilometers for SMMR (November 1978 through June 1987), 0.31 million square kilometers for SSM/I (July 1987 through December 2013), and 0.029 million square kilometers for SSMIS (January 2008 to present). Therefore, there is a discontinuity in the "area" data values in this file at the June/July 1987 boundary and at the December 2007/January 2008 boundary. which does not really make sense to me. Is the irregularity on "area" around the dec 2007/Jan 2008 boundary, or around December 2013 ? Or both ? I'm confused. Anyone know what's going on here ?
Thank you Neven, for again a great overview of the early melting season. I share your conservative view that the low extent (especially from IJIS) is probably not a significant indicator. Low ice "extent" and average "area" suggests that the ice pack is compact and does not yet expose much water (melting ponds) in the ice pack, and thus amplification is limited. Thus,assuming "average" summer weather, I believe that the early melting start may taper off over the next month and get more in line with extent numbers of 2011 and alike. But the Arctic is notoriously unpredictable, and certainly has a few tricks up her sleeve. Also, a big THANK YOU to Chris Reynolds, for the May summary post : What a clear and wonderful read, explaining so comprehensively how the air pressure differences over the Arctic affect regional melting patterns. Hats off !
navegante said Rob, your comment seems to imply that a two-state, hysteretic, irreversible Arctic requires one of the two states to be year-round ice free. Cannot this state be just a seasonally ice free Arctic? Yes, that is true. A two-state, hysteretic Arctic does not need it's second state to be year-round ice free. I only gave that as an example of bifurcation. A more accurate description was made by Chris Reynolds, above : However with Bifurcations often comes rapid transitions. This paper suggests that this is not likely. although, while we witness the reduction of Arctic sea ice over the years, it may be difficult to judge afterwards if it was due to transition to a bifurcation state, or simply a rapid change due to underestimation of positive feedback factors. Either way, the real issue is in my opinion that even GCMs still appear to underestimate Arctic sea ice decline :
Either way, based on SIPN prediction June-Sept methods that have 500 k km^2 SD or less, the 2013 and 2014 Arctic summers were COLD compared to the long term "trend" line. Including my method that uses NH snow cover as a predictor.
Jim said I wondered if I might idly enquire what your next SIPN prediction might say? Short answer : By June 3 or 4 Rutgers will publish their NH snow numbers for May, and that's when I can publish a prediction. The long answer : For starters, I owe Slater et al an apology. Their 2014 prediction, based on melting pond data from in early May, came through while my, and many other, predictions that were smack in the middle of the pack ended up almost 2 SD's off. However, I noticed that the standard deviation of the September extent predictions by the SIPN contributors (including me) is stubbornly close to 500 k km^2 or more, no matter WHICH method was used. Which puts Hamilton's Gompertz fitting method at the same accuracy as Slater et al's early May melting pond method (both of which reported a 500 k km^2 standard deviation for the June->Sept prediction). My own method (which relies on snow cover in the NH) obtains a better SD, but turned out not serve as a good predictor for the 2013 and 2014 melting season. Which is somewhat concerning, since it suggests that (1) physical effects are less important than summer weather, but (2) the long term down trend is indisputable and does not depend much on the weather in summer. And I'm not even sure if that is a paradox simply an indication that Arctic summer weather create a 500 k km^2 SD on June to Sept sea ice extent.
Neven, I'm sorry that I just jump in here with a summary of a scientific paper after a year of absence. I've been busy fighting denial on (shorter term) issues like Keystone XL, Canadian tar sands, MH17 and Russian aggression in the Ukraine.
Regarding the paper by Wagner and Eisenman, here : The issue addressed in this paper is if there is a "bifurcation" point in Arctic sea ice loss, meaning that Arctic sea ice could appear in two states given a certain amount of climate "forcing". Not inconceivable given the fact that sea ice reflects sunlight and thus keeps itself in place, and thus that if sea ice is gone, that a state of dark ocean could be equally likely. If that is the case, then for example, will Arctic sea could "flip" over from summer sea ice to virtually ice free all year around, and we may not be able to regain Arctic sea ice after it is gone. Such "bifurcation" behavior are apparent in SCMs (single column models) and EBMs (energy balance models) but interestingly NOT in the big GCMs (General Circulation Models) that IPCC uses. This Wagner and Eisenman paper shows that if you take heat latitudial heat transport and seasonal variation into account, that the "biforcation" behavior of the SCM and EBM models disappears. So essentially the paper surrenders to GCMs as the more realistic way in which Arctic Sea Ice would develop under increased GHG forcing. Meanwhile the paper suggests that seasonally ice free conditions in the Arctic may occur at lower temperature increase than previously thought. Which means that the GCMs are probably right that Arctic sea ice will not flip over to year-around ice free state any time soon, but they may reach seasonally ice free state sooner than expected. That is a conclusion that seems to be sustained by simply comparing Arctic sea ice development compared to these GCMs :
Chris Reynolds said : Greenland 500mb GPH is back up to similar levels as the 2007 to 2012 years. Also the 'summer pattern' correlation makes 2014 look like a 2007 to 2012 year. But the temperature for JJA was around the same as 2013, and well below the typical for summer in the 2000s. With that statement, Chris, I think you perfectly summarized the issue this year. Why were the temperatures in the Arctic at 2013 level this summer while pressure data suggest it should be a 2007 or 2012 year ? There are good responses to that from various posters above, but no conclusive explanations. Did anyone look at the stratosphere ? Or may it be a combination of factors (small melting ponds in May, compact ice etc) ? Either way, remember that all we are trying to do is finding the 'trend' line by putting bounds on summer variability...
Toggle Commented Sep 10, 2014 on PIOMAS September 2014 at Arctic Sea Ice
If I can make one suggestion, Neven ? Create one forum entry for comments that contain ad hominems and insults (your call). Just dump the whole comment in there, and replace the comment with a pointer to the forum entry. Then create a forum entry for a known myth, and drop each comment that advertises that myth into that thread. That way, you remove all off-topic distractions on your main blog, but posters still have their comments available and other people can still comment on them. Away from your main blog threads. All of this at your discretion.
Toggle Commented Sep 10, 2014 on PIOMAS September 2014 at Arctic Sea Ice
Neven, I'm not sure if I'm going out on a limb here, but with the creation of the forum, it seems that you have diverted the best, and most serious scientific crowd sourcing work to the sideline, only to leave your main blog open to abuse by septics.
Toggle Commented Sep 10, 2014 on PIOMAS September 2014 at Arctic Sea Ice
Kristian Fredriksson said What is wrong with this study? One or two arguments please. This study finds that from the Himalayan glaciers they analyzed, 248 are receding while only 18 are expanding. No argument there, since it is consistent with prior research. The problem is with the 1752 glaciers that the report as "stable". Note that 1453 of these (a whopping 83 %) are smaller than 3 km^2, and 914 of these (about half) is even smaller than 1 km^2. Now remember there spacial resolution is 25 meter or so. Take your average 1 km^2 glacier of 100 meter wide and 10 km long, and you find that these small glaciers would have to have melted 25 % over the past 10 year to be classified as "receding". In other words, the bulk of the glaciers in this study should be discarded, since the study's resolution is way too high, as actually the authors themselves admit (sort of). That is ONE argument against this study. There are many more questions you should ask about this paper, many of which YOU should be able to answer yourself if you would just : DO YOUR HOMEWORK !
Toggle Commented Sep 9, 2014 on PIOMAS September 2014 at Arctic Sea Ice
BDS and planet 8788, here is summer sea ice extent by Walsch and Chapman : which puts your argument of an uptick from 1974 to 1979 into some well deserved perspective. Any other fake skeptical arguments you want to bring up ?
Toggle Commented Sep 9, 2014 on PIOMAS September 2014 at Arctic Sea Ice
In spite of record highs in BC and the Canadian NWT, with massive forest fires in Siberia, a strangely acting jet stream, which is causing a unique event which could be called "Fram import" (MYI from the Greenland sea blowing back into the Arctic Ocean), a Trans Polar drift going opposite direction as usual, and what appear to be sustained high density ice in the Arctic make me now think that maybe this is going to be a good year for Arctic sea ice ! I'd feel a bit more comfortable if the pattern this year can be explained as a 'return to trend' rather than a fluke weather event...
Toggle Commented Jul 22, 2014 on ASI 2014 update 5: low times at Arctic Sea Ice
Blizzard_of_Oz, thanks. I think your method (using ice concentration) is important and interesting, especially for short-term (<50 day) forecasts, but most of all for the insight it gives on the effect of melting ponds and fragmented ice in the melting ice margin. Especially, I found the lower-left graph in your poster intriguing: Your graph suggests that there is a 50% chance that a pixel with 62% ice concentration on July 27 will reduce to 15% ice concentration by September 15. I may be going our on a limb here, but I think that tells something about the ice thickness in the ice margin. 62% ice concentration on July 27 means 38% 'dark' area within the pixel reduced to 15% by September 15 means an average 'dark' area of (85+38)/2=61.5% 'dark' melting out 85-38=47% of the ice in place. From July 27 until September 15, that pixel will receive something like 300 MJ/m^2 of solar energy (ask me about that). 61.5% of 300 MJ/m^2 is 185MJ/m^2, which then melts 47% of the ice. With energy to melt 1 ton of ice set at 330MJ, the ice in the margin that melted out must have been about 185MJ/330MJ/0.47=1.2 meter thick on July 27. More importantly, your graph suggests suggests starting concentration in 2002 of 55% (instead of 62% in 2012), which implies (85+45)/2=65% 'dark' area melting 85-45=40% of the ice (0.65*300MJ)/330MJ/(0.4)=1.48 meter in 2002. With all the inaccuracies of this 'back-of-the-envelope' calculation, these numbers are consistent with PIOMAS and other estimates of ice thickness such as Neven's volume over area, so it seems to me that there is a case to be made that your graph adds evidence to reduced (mostly FYI) thickness. Specifically that ice in the margin (mostly FYI) reduced in thickness from 1.48 meters in 2002 to 1.2 meters in 2012... Or am I way off now ?
Toggle Commented Jul 16, 2014 on ASI 2014 update 5: low times at Arctic Sea Ice
Andrew, thank you for posting here. In your submission to the June SIPN report, you mention that your method (using ice concentration maps) has skill over the 50 period projection time frame, but skill drops below 0 for the September outlook. Can you please explain a bit more about the accuracy (and skill) of your method for periods shorter than 50 days ? For example, does the skill of your method (using ice concentration maps) improve for shorter periods (such as 30, 20 or 10 days) and if so, what is the period for best skill of your method ? And over that period, which 'sensitivity' do you find ? (How many km^2 of ice melt out over that period for 1 km^2 of reduction in ice concentration).
Toggle Commented Jul 15, 2014 on ASI 2014 update 5: low times at Arctic Sea Ice
Ostepop said : There will be no summer without arctic ice in our lifetimes or in the experience of several generations from now. I wish I could share your optimism (or should I say opportunism?). I'd be more comfortable with your projection (that "The bounceback is a reality" and "it will continue.") if our Arctic start showing the 6-7 million km^2 September minima that our models estimated for 2014. even though these models still project ice free summers in our lifetime.
Toggle Commented Jul 14, 2014 on ASI 2014 update 5: low times at Arctic Sea Ice
Thanks Chris, On the forum, you mention that you withdraw that projection, due to an error. Do you have a corrected projection based on PIOMAS data ? Also, you mention that due to time constraints you are "on the verge of retiring my blog". Let me just say that this would be a significant loss for us ice watchers. Your thoughtful insights here on ASI sustained by evidence on your blog are an inspiration and a valuable resource for all of us. As the Arctic never stops to amaze, I hope you can find time to continue in the discussions with your valuable insights and perspective.
Toggle Commented Jul 10, 2014 on ASI 2014 update 4: high times at Arctic Sea Ice
lodger, thanks. For starters, this correlation is not just with May snow cover. It includes March, April, May and June snow cover, as well as a factor of (June-extent minus June-area) which I believe represents melting ponds and polynia in June. I did not find any correlation between snow cover further back (such as Feb) and Sept ice extent. Correlation starts in March, and grows until June. Allow me a day or two to write down the details (which I will submit to SIPN for the July report), since I only have a few minutes right now. And thank you for the link to Lemke et al. I'll read it.