tag:blogger.com,1999:blog-1367053740188758246.post5949967341654458486..comments2023-06-21T00:39:34.443-07:00Comments on Dosbat: Open Water Formation Efficiency Part 2.Chris Reynoldshttp://www.blogger.com/profile/16843133350978717556noreply@blogger.comBlogger8125tag:blogger.com,1999:blog-1367053740188758246.post-40427756834602799342013-02-15T05:57:50.487-08:002013-02-15T05:57:50.487-08:00I am probably just not getting my heat around OWFE...I am probably just not getting my heat around OWFE%.<br /><br />I think your division of NH into 3 areas is reasonable as the area go all the way to North Pole.<br /><br />If you tried applying measure to a region like Kara which had 40cm thickness and reached 100% open water the OWFE% calculates to 250%. <br /><br />What does 250% mean???<br /><br />If we could actually melt 1.5m of ice then perhaps 375% might be a more sensible answer.<br /><br />It seems the measure is not appropriate for areas that completely melt out. Maybe the measure should only be used to compare the same areas at different date. So comparing different area is a bit dodgy? <br /><br />I am struggling to work out what sort of measure I am grasping for.<br /><br />I am probably just not getting my head around it and this is a problem with me rather than a problem for you.crandleshttps://www.blogger.com/profile/15181530527401007161noreply@blogger.comtag:blogger.com,1999:blog-1367053740188758246.post-7507335256940636262013-02-15T04:41:21.236-08:002013-02-15T04:41:21.236-08:00Yep, fine. Got today off but still not firing on a...Yep, fine. Got today off but still not firing on all cylinders. Leaving aside the OWFE issue for here, I see you've been moving forward with your thickness/melt vulnerability idea. I'll 'meet you' <a href="http://neven1.typepad.com/blog/2013/02/piomas-february-2013.html?cid=6a0133f03a1e37970b017ee8848abc970d#comment-6a0133f03a1e37970b017ee8848abc970d" rel="nofollow">over at Neven's</a> to discuss that.Chris Reynoldshttps://www.blogger.com/profile/16843133350978717556noreply@blogger.comtag:blogger.com,1999:blog-1367053740188758246.post-1130090723125992862013-02-14T14:34:55.867-08:002013-02-14T14:34:55.867-08:00No rush. Not sure my thoughts are going anywhere e...No rush. Not sure my thoughts are going anywhere either on OWFE measure or the decile analysis I have been trying.<br /><br />4am start and 12 hour day are tiring, hope it went OK.crandleshttps://www.blogger.com/profile/15181530527401007161noreply@blogger.comtag:blogger.com,1999:blog-1367053740188758246.post-38710427708053057542013-02-13T11:53:25.714-08:002013-02-13T11:53:25.714-08:00Crandles,
Sorry but I've just had a call and ...Crandles,<br /><br />Sorry but I've just had a call and need to start work tomorrow at 0400, so I'm going to have a night off tonight and get back to you tomorrow.<br /><br />Isn't Wipneus's spreadsheet of grid areas in the correct order? If not I can easily copy Rob's into the necessary order.<br /><br />PS tried out what I suggested re referencing melt to April thickness, not what I was expecting.<br />Chris Reynoldshttps://www.blogger.com/profile/16843133350978717556noreply@blogger.comtag:blogger.com,1999:blog-1367053740188758246.post-86955958043230444372013-02-13T09:41:51.131-08:002013-02-13T09:41:51.131-08:00I wouldn't get too hung up on the 100% 1m issu...I wouldn't get too hung up on the 100% 1m issue. It's not mentioned in the paper but I can't see any good reason why one couldn't have a short period at the end of the summer without ice at average thickness (April) of 1.5m and average thinning of 1.5m leading to open water of 100% i.e. OWFE = 100/150 = 0.666...<br /><br />In that case the scatter plot wouldn't reach 1 before 100% open water occurred.<br /><br />Actually there may be a good reason - thicker ice inside the pack and more thinning at the edges. Another that's just occurred to me is that we're talking about monthly averages, not just a few days of 100% OW.<br /><br />This is why I did my first post using %OW as opposed to Holland et al's defined OWFE.<br /><br />I've just been so concerned I might have screwed up again that I checked the source paper: They state - "defined as the percent open water formation<br />per cm of ice melt over the melt season"<br /><br />Perhaps a better metric would be to replace cm of melt over the melt season with 'melt over the melt season expressed as a percentage of April thickness'. Off the top of my head I think that might consistently normalised to zero. But should keep the general form of the scatter.<br /><br />As usual you get complex very fast, I've just got in from work and need a meal and a rest before considering the rest of what you say.Chris Reynoldshttps://www.blogger.com/profile/16843133350978717556noreply@blogger.comtag:blogger.com,1999:blog-1367053740188758246.post-2263243394776933152013-02-13T05:58:33.864-08:002013-02-13T05:58:33.864-08:00Correction in 2012 there are 11245 cell with Sept ...Correction in 2012 there are 11245 cell with Sept thickness less than 10cm.<br /><br />By deciles (1653 cells) 2012 thinning was:<br /><br />Apr Thick, Reduction, Sept Thick<br />0.058553478, 0.058553478, 0<br />0.572480337, 0.572480337, 0<br />1.122923139, 1.122923139, 0<br />1.601478836, 1.601478836, 0<br />2.087651403, 2.087651402, 4.43247E-10<br />1.317933088, 1.317448419, 0.000484669<br />1.548861945, 1.492142027, 0.056719918<br />1.930705475, 1.470039892, 0.460665583<br />2.220763057, 1.142817656, 1.077945402<br />2.895054369, 0.961668537, 1.933385831<br /><br />Maximum thinning decile has over 2m of melt! If the pack was all 2m thick would it all melt? I don't think so, because those will be cells that get to be on the edge of the pack and get melted at a fast rate. Also some cells will be related to transport out of ice. Does that affect as much as 10% of cells? Hmm not sure. Even if there is a decile of transport then the next decile going to 0 thickness shows 1.6m melt. That is a substantial thickness that can melt out.<br /><br /><br />3m ice suffers less than 1m of thinning. Again some (but not as much) could be transport but more likely the 3m ice is MYI that is difficult to melt as well as not being thin to reduce albedo as it gets thinner. Still get 1m of thinning compared to 1.6m so it isn't hugely harder to melt.crandleshttps://www.blogger.com/profile/15181530527401007161noreply@blogger.comtag:blogger.com,1999:blog-1367053740188758246.post-5471156319480984462013-02-13T04:46:02.776-08:002013-02-13T04:46:02.776-08:00Not sure if these numbers are interesting: (hope f...Not sure if these numbers are interesting: (hope formatting isn't too bad)<br /><br />Year 2010, Apr Avg Thick, Reduction, Sept Avg Thick, No cells<br />Sept lt 10cm, 1.138800989, 1.133596751, 0.005204238, 10768<br />Sept gt 10cm, 2.251138011, 1.230686617, 1.020451395, 6731<br /><br />Year 2012, Apr Avg Thick, Reduction, Sept Avg Thick, No cells<br />Sept lt 10cm, 1.174049807, 1.16953493, 0.004514878, 12245<br />Sept gt 10cm, 2.304140099, 1.210651578, 1.093488521, 5289<br /><br />The thinning of cells going to 0 thickness seems surpisingly similar to thinning of cells that retain some thickness. I guess that is just co-incidence with thinning of cells in center of (MYI?) pack and outside edges being lower than maximum thinning cells and it just happens to average out.<br /><br />BTW, Have you got a csv file of latest cell areas in same order as heff files?crandleshttps://www.blogger.com/profile/15181530527401007161noreply@blogger.comtag:blogger.com,1999:blog-1367053740188758246.post-43292980728539454422013-02-13T02:46:11.888-08:002013-02-13T02:46:11.888-08:00If the thickness is a meter then for 100 cm of mel...If the thickness is a meter then for 100 cm of melt, OW is going to result so it is not surprising that OWFE reaches 100% at 1m.<br /><br />I am wondering if this is a sensible measure.<br /><br />Last year we had 18.6 K Km^3 volume reduction and area at time of volume maximum was 12.93 m km^3. This implies average thickness melt of 1.44m. As there is lots of thin areas with less than 1.44m thickness at the volume max, then the melt season would seem likely to melt any cell with 1.5m or less ice.<br /><br />Therefore should we have some form of OWFE measure that approaches 100% at a thickness of 1.5m rather than at 1m?<br /><br /><br />If we calculated sum of ((lower of Piomas thickness and 1.5)*cell area) for April of each year. How would that differ from volume reduction in that year?<br /><br />Perhaps find the figure which if used instead of 1.5 makes the volume reduction correct and see how that figure changes.<br /><br />Could this better predict the volume reduction?crandleshttps://www.blogger.com/profile/15181530527401007161noreply@blogger.com