Please support this Forum and Neven's Blog

Author Topic: Does the Volume Midway Point have predictive value?  (Read 338 times)

Dave C

  • ASIF Lurker
  • Posts: 37
    • View Profile
Does the Volume Midway Point have predictive value?
« on: July 09, 2017, 02:29:27 PM »
Does the midway point for ice melting have predictive value? Based on recent trends it seems that it does.

----------Max-----Date----Half-----Date-----Min-----Date-----Freeze---Melt-----%Melt---Max/AprExt
2001---27.75---109-----19.96---186-----12.18---260------16.80----15.57----56.1----1.86
2002---27.54---110-----19.17---187-----10.79---260------15.36----16.75----60.8----1.91   
2003---27.32---105-----18.78---186-----10.24---253------16.53----17.08----62.5----1.86     
2004---25.81---115-----17.85---190-------9.88---254------15.57----15.93----61.7----1.82   
2005---26.18---113-----17.67---185-------9.16---260------16.30----17.02----65.0----1.85     
2006---25.19---105-----17.09---185-------8.99---265------16.03----16.20----64.3----1.79           
-------------------------------------Transition point---------------------------------
2007---23.87-----98-----15.17---183-------6.46---264------14.88----17.41----72.9----1.71       
2008---25.16---108-----16.12---187-------7.07---262------18.70----18.09----71.9----1.74           
2009---25.07---103-----15.96---183-------6.84---255------18.00----18.23----72.7----1.71 
-------------------------------------Transition point---------------------------------
2010---24.28---100-----14.43---177-------4.58---258------17.44----19.70----81.1----1.64
2011---22.68---109-----13.49---179-------4.30---253------18.10----18.38----81.0----1.59 
2012---23.37---114-----13.52---176-------3.67---261------19.07----19.70----84.3----1.58
2013---23.33---112-----14.36---180-------5.39---257------19.66----17.94----76.9----1.62     
2014---23.12---105-----14.97---180-------6.81---261------17.73----16.31----70.5----1.62   
2015---24.39---111-----15.04---182-------5.68---255------17.58----18.71----76.7----1.75   
2016---22.72---112-----13.56---180-------4.40---250------17.04----18.32----80.6----1.64
2017---20.78---108----------------------------------------------16.38-------------------------1.50

"Half" is the date on which half of the ice melts each year. The beginning and ending date of each melt season seem to be mostly random. The midway point has a strong trend though. As the maximum drops this correlates with earlier midway points. Increased ice melt also correlates with earlier midpoints. This year has competing trends- a very low maximum but also unusually low melt so far.
Based on the Max vs Halfway function the midway day should be 176. This would result in a stunningly high minimum this year. However, since melt is low then the midway day might shift forward. This is the projected minimum for each midway day and the implied midway date for each melt total based on the Melt vs Halfway function.
Halfway Day Possibilities-
Mid----Value----Implied melt--Implied min--Melt/Midway projected midway
176---13.353---14.85---------5.93----------189     
177---13.087---15.39---------5.39----------188       
178---12.823---15.91---------4.87----------187     
179---12.599---16.36---------4.42----------186       
180---12.377---16.81---------3.97----------185   
181---12.164---17.23---------3.55----------184

Deviation from max vs midway function-
176-0
177-1
178-2
179-3
180-4
181-5
Deviation from melt vs midway function-
176-13
177-11
178-9
179-7
180-5
181-3

The best compromise between deviations is day 180. Anything beyond day 181 is implausible. So if 180 is the midway point then the volume record will not be broken and most people would be surprised- but at least it would be a plausibly low minimum. However, is each function equally predictive? Why does each relation hold in reality?

The cause for the max vs midway correlation seems simple. As ice volume declines the ice is thinner. Ice melt will increasingly be driven by the summer solstice which occurs on day 172. Obviously that reason strongly holds this year.
But why does the melt vs midway correlation exist? Does it occur because when there is low melt in the first half of the season that means that there is more ice to melt later in the summer? Or that there is more ice at lower latitudes that has more sun/heat to melt? These explanations seem plausible. But they also don't apply at all to 2017. If only the max function applies this year then the ice volume minimum would be shockingly high. Of course there could be other reasons why these correlations exist. And it hardly needs to be stated that when dealing with this few data points that the trends are hardly iron laws.
But as of now my prediction is that the volume minimum will not be broken this year, and that the minimum has a decent chance of being far higher than most would expect.

Other questions-
Will a reduction in melt start to happen as the max gets lower? This year is so much lower that this year could be the first year that this effect shows up.
Has ice volume plateaued since 2010? There doesn't seem to be much of a trend since then. 
Can 2017 beat the recent record low melt set in 2014? The max vs midway relation says it will. Most other numbers say that it will not.

greatdying2

  • ASIF Citizen
  • Posts: 370
    • View Profile
Re: Does the Volume Midway Point have predictive value?
« Reply #1 on: July 09, 2017, 08:05:13 PM »
Hi Dave. Interesting idea, but I'm having a bit of trouble following. Any chance you could put this in graph form?