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
CNN said ""142 additional inches of rain"" I'm pretty sure that is 142 mm of rain.
John said : What if positive AO (increased cloudiness) during Oct-Dec and negative AO (clear skies) during Apr-June would make enough impact on solar radiation on surface temperatures to have any impact on summer Arctic sea ice melting? To see if there is something to this argument (of winter/spring AO affecting Sea ice minimum), I decided to run your (2006-2015) AOwinter - AOspring numbers through linear regression against the September ice extent numbers. The result is interesting. The correlation R = -0.415, which is not particularly good (only explains 17 % of the variance in Sept ice extent), but it is not bad either (it is better than linear extrapolation). Also the sign is as expected : High AO in winter followed by low AO in spring cause a decrease in Sept sea ice minimum. That is encouraging. On the other hand, to be frank, your time period (of just 10 years) is too short. Linear regression takes 2 data points away, so with 8 data points there is a significant risk of statistical "over-fitting" (so the 17% is optimistic). So predictive value of this AO formula is not strong, but may serve as an loose indicator of what's to come the next summer. P.S. Your AO formula would have predicted 4.45 for 2012 and predicts 4.50 for 2016. Standard deviation 560 k km^2.
Toggle Commented 2 days ago on 2016 Arctic cyclone, update 2 at Arctic Sea Ice
dominik said How does PIOMAS deal with such a dispersed and at the same time compacted ice mass? That is a very good question. Anyone with more knowledge about this please correct me, but from what I understand, PIOMAS adjusts their "ice thickness" at any particular grid point by assimilating the ice concentration observation with the modeled ice thickness. So if one gridpoint has modeled ice thickness of 1 meter, but the ice concentration observation is 50%, then they present the ice as having 0.5 meter thick ice. That way, ice volume calculations are following ice concentration quite nicely and do not show unrealistically thick ice in the marginal ice zone.
Toggle Commented 3 days ago on 2016 Arctic cyclone, update 3 at Arctic Sea Ice
John, thank you for your elaborate explanation of the causality pattern (winter AO causing low snow cover causing summer AO causing summer cyclones causing ice extent changes), and thank you for the scientific papers that sustain the individual steps. I find this reasoning plausible, but because of the long chain of effects, I wonder how 'strong' the correlation is between the starting point (winter AO) and the end point (summer minimum ice extent). If it is strong, then the effect you describe should be very useful as a predictor for summer ice extent, and thus be of interest to ARCUS/SIPN. Also one additional question : Is there any indication that the winter AO is changing over the past decades ? Or is it simply random ?
Toggle Commented 3 days ago on 2016 Arctic cyclone, update 2 at Arctic Sea Ice
Cato said : the persistence of a very strong HP at the high latitudes of the Pacific ocean has forced LP systems to move to the Arctic I suspected that that "ridiculously resilient ridge" over the North Pacific had something to do with it all, but is there any evidence that such a HP zone over the North Pacific creates LP systems over the Arctic ?
Toggle Commented 4 days ago on 2016 Arctic cyclone, update 1 at Arctic Sea Ice
Hi John, You make a valid argument, and it may indeed be that due to land snow cover decline in summer that summer cyclones on the Arctic would become more commonplace. Two questions about that : 1) Is there any evidence that cyclones in summer became more common when ice extent (or land snow cover) was low ? 2) If 1) is true, then did these cyclones "help preserve the sea ice" as you assert ?
Toggle Commented 4 days ago on 2016 Arctic cyclone, update 2 at Arctic Sea Ice
NIce summary, Neven. It still amazes me that despite the cyclones, and despite the flying start, we still can't quite rank where 2016 will end up in September. It may come down to the question of if the "arm" of low concentration ice towards the ESS will melt out or not.
Toggle Commented 4 days ago on 2016 Arctic cyclone, update 3 at Arctic Sea Ice
Neven said I'll be doing a piece on the Walsh et al. paper around this year's minimum. That's great, Neven, thank you ! Remember that in the discussion with Diablo, we found essentially three issues with the otherwise fine Walsh reconstruction: 1) The Kelly fields seemed to be misplaced by one month (August shows July Kelly fields, and July shows June Kelly fields). This results in a significant high-bias for the 1935-1952 reconstruction. 2) The Walsh & Johnson source has a high bias, which results in a high-bias of the 1953-1979 reconstruction. 3) The DMI source in Walsh reconstruction does not account for the "ice edge" that is depicted as a red lines in the DMI charts. This results in a small high-bias caused by ice fill-in of neighboring pixels that really should be open water. Issue (3) was never resolved, but it is most likely not significant. Issue (2) was quite easily resolved by adjusting for the ice concentration of the Walsh & Johnson source. Issue (1) is much more difficult, since when we remove the Kelly fields, we need to create an alternative "fill-in" algorithm, which is what I worked on (results posted above) in May. I left that work at the end of May, when I introduced temporal fill-in into my algorithm. That resulted into reconstructions for August and September, but unfortunately the difference between August and September for many of the years was unrealistically large (more than 1 million km^2 for many of the years in the 1935-1952 period). I suspect that the "open water" observation sensitivity that I reported about earlier is playing a role here, but the 2016 melting season got in the way, but I did not investigate the issue further. Now that you plan to post about the Walsh paper, it is time to pick up that work where I left it in May :o)
The surplus heat needed to explain the loss of Arctic sea ice during the past few decades is on the order of 1 W/m2. This finding by Kwok and Untersteiner suggests that the Arctic is more sensitive to weather pertubations than we would normally assume. It also suggests that (all things included) Arctic sea ice is a very good measure of global climate forcing.
Thanks Neven, for this summary. I find it amazing that (on Aug 21) it is still not obvious where 2016 will end up in the ranking (2nd or 3rd).
Regarding the 4.1 projection : are you predicting the lowest daily volume or the Sept average extent? I'm projecting Sept monthly average, using June data. Here are the details of the method : Note that the formula given is slightly different than the formula I used in my first blog comment back in 2013. Back then, I used the 1995-2012 period to calibrate, which resulted in a 4.5 M km^2 prediction for Sept 2013. Now I use the 1992-2015 period to calibrate, which performs better for the upside outliers and would have predicted 5.1 for 2013. This new formula is the same I used last year, and at that time it was spot-up (4.6). Overall, here is what this method would have produced over the past 24 years :
Toggle Commented 7 days ago on 2016 Arctic cyclone, update 1 at Arctic Sea Ice
Regarding the predictive method (using ice area and extent and land snow cover), it is interesting to see how that evolved this year : - At the start of May, it was pointing at 3.5 M km^2 (albeit with a large standard deviation), while - at the start of June, the data suggested 3.8 (still with a large standard deviation), and - at the start of July, the prediction was 4.1 M km^2, with a nice and tight standard deviation (see the July SIPN report). And note that due to the reducing standard deviation, the actual probability of the September minimum ending up around 4.1 actually did not reduce very much over the months. This progression of the prediction suggests what we all know already : That despite a flying start, the melting season was not very conducive to melt. It also suggests that influence of weather is clearly present, but limited. After all, the difference between 3.5 and 4.1 is not THAT large. As for the June prediction (of 4.1), I still stand by that. It is the number we would get for 'average' weather over July, August and September. July was clearly negative for melt, August may be positive for melt (stirring up lots of heat from the ocean during this storm). As for September, I think that there are large ice areas that are isolated from the main pack, which will melt out in September, bringing the September minimum rather close to the expected 'average melt' prediction of 4.1. But time will tell.
Toggle Commented Aug 19, 2016 on 2016 Arctic cyclone, update 1 at Arctic Sea Ice
During the GAC 2012, there were two ITP buoys that reported on significant disruption of the halocline below, suggesting massive heat being provided to the bottom of the ice during the storm in an effect called Ekman pumping. This time, since no ITPs are operational in the Arctic at this time, there will be no witnesses of this effect. Which is sad and disappointing that we could not get at least a couple of buoys recording what is about to happen. But IF the ice starts to disappear in flash melts during this significant 2016 storm, you know where is comes from...
Toggle Commented Aug 14, 2016 on ASI 2016 update 5: big cyclone at Arctic Sea Ice
Thanks John, nice that DMI replied to your inquiries. Seems from their notes that the OSISAF extent changes that they applied did not affect the volume numbers in any significant way (at least according to their own assessment). What remains is the differences between PIOMAS and DMI volume charts. Regarding deniers referring to the DMI volume chart, I guess this must be due to DMI using a different reference point (2004-2013 mean) compared to PIOMAS (1979-2015 mean). Yes, that is one thing. But there are more fundamental differences between DMI's Hycom/CICE model and PIOMAS. For example : - DMI volume puts 2016 at this time in the middle of the (2003-2016) pack, while PIOMAS puts 2016 at 3rd place (more in line with the extent ranking). - DMI puts 2007 volume as being lower than 2016. PIOMAS puts 2007 much higher than 2016, reflecting the higher ratio of MYI versus FYI in 2007. - DMI puts ice volume loss from May to end of July as the smallest in the entire 2003-2016 record. We know that 2016's meting season was not conducive to melt, but smallest volume loss in 13 years is simply not plausible.
Toggle Commented Aug 12, 2016 on PIOMAS August 2016 at Arctic Sea Ice
Bill said : Rob, there may be such an example about to appear on Jim Hunt's "Great White Con" blog. I have just asked someone to provide a citation for their claim that, amongst other things, "... the ice volume is now very close to the 30 year average ..." Yes. I see that exact same comment on a WUWT post (by one Frank Lansner) which also argues with DMI volume graphs (completely ignoring PIOMAS). DMI, "the gift that keeps on giving" strikes again. John Christensen, any more info on "They significantly enhanced ice concentration and other algorithms in June this year" and if they retroactively adjusted the data or not ?
Toggle Commented Aug 11, 2016 on PIOMAS August 2016 at Arctic Sea Ice
Chris said : On the DMI volume chart: They significantly enhanced ice concentration and other algorithms in June this year Do you have some more information about that ? I see DMI's volume charts are used in denier circles more and more, to claim all kind of 'recovery' statements. To counterbalance such statements, it would be good if there were some more info on DMI's volume graphs. For example, what kind of ice concentration and other algorithms were enhanced, and how ? Did they retro-actively adjust past years with the same enhancements ? Is there an overview of past year DMI volume data, as processed with the same setting of their algorithms, like there is for PIOMAS ?
Toggle Commented Aug 10, 2016 on PIOMAS August 2016 at Arctic Sea Ice
Bill said Does that mean you're becoming bi-polar? Bill, that is quite ad-hominum.
Toggle Commented Aug 9, 2016 on PIOMAS August 2016 at Arctic Sea Ice
John said : I would also just note that July 2016 saw less volume loss than 7 of the last 10 years from your list above, I noticed that too. Kind of puts the emphasis on Neven's remark that : "I can't escape the feeling that the Arctic sea ice has dodged a bullet this year. " I could not agree more.
Toggle Commented Aug 7, 2016 on PIOMAS August 2016 at Arctic Sea Ice
Cato, Do you have the DMI thickness maps from 2015 and 2012 from around this time ? Would be interesting where these exactly these years had less ice.
Toggle Commented Aug 7, 2016 on PIOMAS August 2016 at Arctic Sea Ice
Neven, about the Olympic opening ceremony, here in California we got to see that as one of the last ones on the planet. Even though the NBC commenters did not say a word about the global warming issue raised during the event, and even though NBC's coverage of the event was interrupted by many car ads and even an Exxon Mobil ad (I kid you not) I still found it inspiring that the event raised awareness for the issue that is central to our efforts here, and proud to be part of the platform you created to share that with the world. Thank you Neven.
Toggle Commented Aug 6, 2016 on PIOMAS August 2016 at Arctic Sea Ice
If 1 Million km^2 disappears from the NSIDC daily extent picture, there is no shame in calling that a "mega melt week". The 942 k drop that we just witnessed came very close to that, despite other graphs and other data out there. Nuf said.
Report from the Forum, by BFTV :,1457.msg85465.html#msg85465 A drop of 188k takes the daily NSIDC extent down to 2nd lowest on record. It also puts the 7 day loss up to 942k, and means that we only have to average a 138k drop over the next 2 days to achieve a mega melt week (>1,000,000km2). That kind of sums up the current developments in Arctic sea ice.
Hi John, NOAA presents a temperature profile over the Arctic which puts the low temps over the remaining ice rather than the LP/HP areas from DMI's graph : Which one is more accurate ?
Just to be clear, I've gone on record to declare that : There is a 92% chance that 2016 will end up in the top-3 (only 2 out of 24 years showed a larger difference between 2011 and this 2016 projection). There is 66% chance that 2016 will be second place (after 2007 but shy of 2012). There is an 8 percent chance of 2016 beating the 2012 record in September.
wayne, I'm sorry. I thought you were talking about land-snow, which held out longer along the Laptev coast this year. Regarding the snow-on-ice graph that you present, it is indeed interesting that it matches the ice-melt pattern this season, at least when comparing the ESS and the Laptev. But if snow cover on ice really would make a substantial difference for ice thickness, should we not see that reflected in the ice-thickness graphs from observations like Cryosat ? Here is that Cryosat graph for March'16 : I do not see much of a difference between the ice thickness in the Laptev and the ESS. If anything, it seems the Laptev (which received the LEAST amount of snow in you graph) has THINNER ice than the ESS. May I suggest that the earlier melt of the ESS (than the Laptev) this year was caused by some quite significant "torch" events in June and July ? I think you even reported on that as well. Long story short, I don't see that "snow cover" on ice has affected the melting season in any significant way this year yet.