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
California
Recent Activity
wayne, as I said before : 80% albedo could be right even if 50% makes it to the ground. Remember that TOA albedo depends on surface albedo, and if that is 50%, and 50% makes it to the ground, then TOA albedo will be 75%. That still does not make your numbers right. You want patience, fine. Let's start all over again : How much of TOA insolation makes it to the surface of the Arctic ocean, wayne ?
Toggle Commented 9 hours ago on PIOMAS February 2017 at Arctic Sea Ice
wayne said ,2 , 20% gets through? Seems clear to me Right. But you made that up all by yourself. Once more, these two maps : https://en.wikipedia.org/wiki/File:Insolation.png show that 50% gets through on average. wayne, I've always given you the benefit of the doubt, and at time stood up for your arguments. But at this point I've had enough. Either you accept the maps above, or you can present better evidence. There is not much in between.
Toggle Commented 9 hours ago on PIOMAS February 2017 at Arctic Sea Ice
wayne, I posted such a nice and concise and important post explaining the strength of the albedo feedback in the Arctic. why are you going over this with absurt arguments ? For example, where do you get your "cloud albedo" factor .2 from ?
Toggle Commented 9 hours ago on PIOMAS February 2017 at Arctic Sea Ice
wayne said : I think a simple map without inscriptions needs to be proof checked.... OK. Once more, the full link is here : https://en.wikipedia.org/wiki/File:Insolation.png The graph at hand (for surface insolation) is from HadCM3. Now, even if you don't like HadCM3, I would be very much interested in your reference to something better. Also, wayne, what the heck is you problem ? Why is this graph/map so unacceptable to you ?
Toggle Commented 10 hours ago on PIOMAS February 2017 at Arctic Sea Ice
You just keep going at this, don't you ? "being 250 w/m2 X .77 atmospheric absorption X .2 cloud albedo X 0.5 sea ice albedo." No wayne, that expression is simply false. For starters, the 250 W/m^2 is wrong (should be 340) Next the 0.2 cloud albedo is wrong (.2 is TOA albedo, NOT cloud albedo). taking 200 W/m2 TOA does not make 100. from 23% atmospheric absorption , 80 % cloud albedo and a generous 50% sea ice albedo. Now you are just rambling.
Toggle Commented 10 hours ago on PIOMAS February 2017 at Arctic Sea Ice
Sorry. An estimate is no longer enough. If you have any better scientific reference than this one : https://en.wikipedia.org/wiki/File:Insolation.png then please let me know. Until then, the rule of the thumb that 50% of TOA insolation makes it to the surface in the Arctic stands. Can we please move on to my calculation and estimate of albedo feedback in the Arctic now ?
Toggle Commented 10 hours ago on PIOMAS February 2017 at Arctic Sea Ice
wayne said Nice maps with hardly a word explanation to them. Need more solid reference. You gotta be kidding me. I'm sorry wayne. We have gone through this two times now (first in the "a-new-arctic-feedback" thread, and now in this). It's very clear that Arctic insolation at the surface is about 50% of TOA insolation. If you have any better estimate, then please post it.
Toggle Commented 10 hours ago on PIOMAS February 2017 at Arctic Sea Ice
wayne said : But I think we are confusing TOA with surface? I don't think we are confusing at all. These two maps : https://upload.wikimedia.org/wikipedia/commons/7/78/Insolation.png show both TOA and surface insolation.
Toggle Commented 10 hours ago on PIOMAS February 2017 at Arctic Sea Ice
wayne, I think I went on record to suggest that the "melt potential" in the Arctic is some 9.5 meters of ice : http://neven1.typepad.com/blog/2017/01/a-new-arctic-feedback-.html?cid=6a0133f03a1e37970b01b8d25ced44970c#comment-6a0133f03a1e37970b01b8d25ced44970c That was based on 100 W/m^2 annual insolation. You can lower that by 10% if we go with the 90 W/m^2 annual insolation on the surface.
Toggle Commented 11 hours ago on PIOMAS February 2017 at Arctic Sea Ice
wayne, man, you are going way off. We just determined from this map : https://en.wikipedia.org/wiki/File:Insolation.png that insolation on the surface is 90 W/m^2 if averaged over 12 months. Over the 6 months that the sun is shining, that is 180 W/m^2 from March till September. Can you take it from there ?
Toggle Commented 11 hours ago on PIOMAS February 2017 at Arctic Sea Ice
wayne, take this graph again : https://en.wikipedia.org/wiki/File:Insolation.png The top map suggests that TOA insolation in the Arctic is some 170 W/m^2. The bottom map suggest that insolation at the surface is some 90 W/m^2. That means that on average, 90/170 = 53% of TOA insolation makes it to the surface in the Arctic.
Toggle Commented 13 hours ago on PIOMAS February 2017 at Arctic Sea Ice
@wayne OK then a transect from March to September gives 260 W/m2 TOA This graph (the top one) https://en.wikipedia.org/wiki/File:Insolation.png suggests about 170 W/m^2 TOA insolation over 12 months, which would be average of about 340 W/m^2 TOA insolation over the 6 months March-September. 50% reaching the ground is too generous, summer time Arctic albedo was measured 80% by NASA. Mmm. 80% albedo could be right even if 50% makes it to the ground. Remember that TOA albedo depends on surface albedo, and if that is 50%, and 50% makes it to the ground, then TOA albedo will be 75%. Now the number makes better sense: 14,038 km3 is the melt potential with a lowered TOA to 250 W/m2. I'm sorry, you lost me. How did you get to that 14,083 km^3 number ?
Toggle Commented 13 hours ago on PIOMAS February 2017 at Arctic Sea Ice
@zebra, I understand. Let's see if this helps. Using this formula : delta (ice) = alpha - beta*(MF) Calculate "MF" for each June between 1992 and 2015. Then calculate 'delta (ice' for each June between 1992 and 2015 as well. That results in what we call a "scatter plot", with "MF" at the x-axis and 'delta (ice)' at the y-axis. Something like this (from the Wiki page on regression analysis) : https://upload.wikimedia.org/wikipedia/commons/3/3a/Linear_regression.svg On that scatter plot we then run simple regression analysis which results in an offset (alpha) and a slope (beta). Does that make sense ?
Toggle Commented 14 hours ago on PIOMAS February 2017 at Arctic Sea Ice
Jim, thanks ! Obuoy 14 appears to have survived the winter ! That's awesome. Let's hope it survives a lot longer, since it is the only buoy in the Arctic with a web cam.
Toggle Commented 14 hours ago on PIOMAS February 2017 at Arctic Sea Ice
zebra, Your request suggests that you don't yet fully understand what I did. So I need some description with words, but I will follow it up with numbers. In the regression formula : september_extent - june_area = alpha + beta * (Melt_Formula) I'm trying to determine the melt between June and September, based on a formula that contains 3 different factors that are available in June. After many regressions, this is the formula that gives the best result (smallest standard deviation) : Melt_Formula = 0.434*june_snowcover - 1.0*(june_extent - june_area) + 0.65*june_area Important to note is that this melt formula is really a 'feedback' factor. If ice melt between June and September were always the same and thus "less(more) ice in June, you will inevitably end up with less(more) ice in September " as you suggest, then the 'beta' factor will be zero and the 'alpha' factor will be simply be the average june-to-september melt amount. But reality is different : If you run linear regression over the 1992-2015 data to find the best fitting parameters, you will find 'alpha' is -5.8538 and 'beta' is 0.36827, an correlation factor R=0.9426 and a standard deviation of the residual of 0.9426. What that means is that sea ice melt between June and September is not constant, but is mostly a result of the various feedback effects of snow cover, 'holes' in the ice, and plain old 'area'. To obtain the prediction for 2016, simply fill in all the june 2016 data into the Melt_formula, multiply by 'beta' ( 0.36827 ) , add 'alpha' (-5.8538) and subtract the june 2016 'area' number from NSIDC. Please do that, and see if you obtain the same number I obtained in the SIPN July report : 2016 September extent projected to be 4.12 M km^2.
Toggle Commented yesterday on PIOMAS February 2017 at Arctic Sea Ice
@wayne What I am interested in is an acceptable-consensus w/m2 insolation value, For the entire Arctic Ocean sun year, A good rule of the thumb is that 50% of TOA insolation makes it to the ground. So the insolation value you are looking for is half of the insolation value from this graph : http://2.bp.blogspot.com/-k_S7N0VlMRg/UH4RNvJ3cjI/AAAAAAAAFjk/lqjQhGqLWOk/s1600/insolation_latitude.gif @zebra, not sure what you want to show with that woodfortrees graph, but it is different from the feedback I calculates. For starters, .42 and .67 year fractions do not correspond to June and September, and what is that 'normalize' function doing ? Is seems to flatten out the trends for both graphs. Please clarify your point.
Toggle Commented 2 days ago on PIOMAS February 2017 at Arctic Sea Ice
Let me also note that the calculations above suggest that if FYI becomes thinner (due to an excessively warm winter) that the albedo feedback in the next summer will be stronger.
Toggle Commented 3 days ago on PIOMAS February 2017 at Arctic Sea Ice
@zebra, yes, the term "albedo amplification" is kind of a misnomer. After all, "albedo" is not amplified here. Better would be "albedo feedback" or even better (as Bill explained) "the component of Arctic Amplification caused by the +ve feedback introduced as a direct result of Albedo reduction". What has captivated me over time is a way to quantify the strength of this albedo feedback, so that we may have an idea on when and how it could cause the first ice free summer in the Arctic. In that regard, Tamino did a series of posts that are very interesting in quantifying the albedo feedback effect of sea ice in the Arctic : https://tamino.wordpress.com/2012/10/01/sea-ice-insolation/ of land snow cover : https://tamino.wordpress.com/2012/10/05/snow-2/ and of ice and snow cover combined : https://tamino.wordpress.com/2012/10/08/snowice-by-request/ Note that he estimates that the albedo feedback effect of sea ice is 0.13 W/m^2 globally, but snow cover reduction is contributing more at 0.21 W/m^2 global climate forcing. The combined 0.34 W/m^2 (even 0.45 W/m^2 if you use a more realistic time filter) is quite substantially adding to global warming over the past few decades. His goal was to estimate the global forcing of reduced ice and snow cover during the melting season, which is different from my goal to estimate the strength of albedo feedback on sea ice retreat. What we really want to know is not the global climate forcing, but how strong the albedo feedback works locally. Like, if there is a hole in the ice in the Arctic in June, then how large will that hole become in September, based on albedo feedback alone. For that, I'd like to present two methods. One is a calculation based on albedo and insolation (that graph I showed above), and one is based on observations. 1) To calculate how much more heat is absorbed by a hole (open water) than by ice, I use the insolation at Top Of Atmosphere (TOA) graph here as a starting point : http://2.bp.blogspot.com/-k_S7N0VlMRg/UH4RNvJ3cjI/AAAAAAAAFjk/lqjQhGqLWOk/s1600/insolation_latitude.gif Note that it peaks at about 500 W/m^2. Now, I use the rule of the thumb that half of that makes it through the clouds, and the albedo different between ice covered and open-water is about 0.6-0.1=0.5, which means that overall TOA albedo should be about 0.25. Tamino in his post here https://tamino.wordpress.com/2012/10/01/sea-ice-insolation/ uses 0.2, which is consistent with Hudson 2011. Now, for a hole in the ice, with that 125 W/m^2 peak (or 100 W/m^2 if you go with Tamino), to estimate how much extra heat is absorbed from June to September, we need to integrate the insolation curve over June to September. That is a Cosine curve from 0 to 90 deg, the integral of which is a Sine of 90deg which is 1. The total energy absorbed between June and September then becomes : 1 * 125 W/m^2 * 3600 * 24 * 91(days) / (2*pi (radians/year) = 156 MJ/m^2. That is enough heat to melt 156 MJ/330,000 = 474 kg of ice. For 1.5 meter thick FYI, that means the hole will get 32 % bigger between June and September. Now, this suggests that the albedo feedback is quite strong, and it will advance the melting of FYI during the melting season, but it also means that there is no imminent danger of a "blue ocean event". 2) Some of you may have followed my estimates based on observation of how albedo feedback specifically about snow cover reduction and ice reduction affects the sea ice minimum in September, for ARCUS Sea Ice Prediction Network (SIPN). My latest assessment is here : https://www.arcus.org/files/sio/25738/sio-2016-july_dekker.pdf Based on these regressions, the best 'fit' (over the past 24 year) appears to be one where 'beta' is 0.368, which means that a 'hole' (extent-area) in the ice would expand (during an average summer) by 36.8 % between June and September. That is quite consistent with the theoretical calculation in (1) above based on albedo and insolation alone. Also note that the best fitting factor for snow reduction over the past 24 years is 0.434, which (with a beta of 0.368) suggests that every m^2 of land snow lost in June will result in 0.434*0.368=0.16 m^2 of sea ice loss. 160,000 km^2 sea ice lost in September for every 1 M km^2 of land snow lost in June. Once again, these numbers (and Tamino's assessment) suggest that the albedo feedback is quite strong, and significantly contributes to global warming and sea ice retreat over the past decades, but it is not 'instable' in the sense that it does not cause a blue Arctic all by itself.
Toggle Commented 3 days ago on PIOMAS February 2017 at Arctic Sea Ice
Glenn said If you assume only ~200 W/m2 insolation during September (a shot in the dark on my part, I couldn't find it through a quick google search), then a single hour of blue ocean at 90 degrees N would result in over 4 PWh of energy injected into the waters of the arctic. Insolation in September over the Arctic is about 50 W/m^2 : http://2.bp.blogspot.com/-k_S7N0VlMRg/UH4RNvJ3cjI/AAAAAAAAFjk/lqjQhGqLWOk/s1600/insolation_latitude.gif It is clear from this graph that the insolation during summer is much more important than insolation in September. Needless to say that albedo amplification is much stronger during the summer months than it is during September.
Toggle Commented 5 days ago on PIOMAS February 2017 at Arctic Sea Ice
Forget about 10 million devices in the Arctic. I'd be VERY happy if we could have just 1000 buoys working in the Arctic measuring the changes that are happening. Right now, there are only a few dozen buoys in the Arctic, and only one with a camera (obuoy14; oops, that one died in November), and one or two that measure below the ice (ITP98).
Toggle Commented Feb 13, 2017 on PIOMAS February 2017 at Arctic Sea Ice
Elisee said I'm sorry if I've been a little too vague and overgeneralized here, but maybe its because I've detected a bit too much certainty, too much trees and not enough forest. Yes. You are very vague, and this latest remark is even more so. Can you specify what exactly you have 'detected' with exactly which remarks here in this thread ? Elisee said Basically, we're faced with a problem where we don't know all the parameters, much less all the solutions. I disagree. We know the parameters good enough to know what the problem is, and we already know the solutions too.
Toggle Commented Feb 10, 2017 on A new Arctic feedback (?) at Arctic Sea Ice
"Jan 2016 - Jan 2017 had a 3.4 ppm(v) increase." Ouch.
Toggle Commented Feb 9, 2017 on PIOMAS February 2017 at Arctic Sea Ice
Sorry, that is 3 GJ/m^2.
Toggle Commented Feb 9, 2017 on A new Arctic feedback (?) at Arctic Sea Ice
@ Kevin In an otherwise excellent post, there is another oddity : If we diffuse that over the top 200m we quickly drop the actual increase in ocean temperature down to a 0.6 degree. Based on insolation, the Arctic receives some 3 GJ per melting season. If that is all put to warming the ocean, we are looking at 37 C increase down to a depth of 20 meters (the mixed layer). If diffused down to 200 m, that is still 3.7 C over the season, not the 0.6 C you assert.
Toggle Commented Feb 9, 2017 on A new Arctic feedback (?) at Arctic Sea Ice
Elisee, that "pretty effective self-moderation" is called Stefan–Boltzmann law. https://en.wikipedia.org/wiki/Black-body_radiation Which keeps the temperature of a planet in check, and vary only minor with changes in GHG forcing. And then there are feedbacks. But I guess you already knew all that.
Toggle Commented Feb 9, 2017 on A new Arctic feedback (?) at Arctic Sea Ice