Support the Arctic Sea Ice Forum and Blog

Author Topic: System for evaluating melting seasons ??  (Read 653 times)

Rich

  • Guest
System for evaluating melting seasons ??
« on: June 29, 2019, 03:55:38 AM »
I'm thinking out loud here about a rating system for melt seasons and I'll use 2019 as an example. I'll consider it similar to evaluating a baseball player where a guy like Mike Trout is considered the best because he does so many things well, but is not necessarily the best in any single attribute

Ultimately, ice melts as a new result of it's intersection with heat. There are a lot of ways that can come about.

1. Atmospheric Heat (2019 score 8.5 / 10)

Plenty of early season heat. Recent scorching heat off Siberia. Consistency in terms of maintaining a positive anomaly throughout.

2. Ocean Heat (10 / 10)

The Pacific side is off the charts and the Atlantic is sporting a 0.5C positive anomaly.

3. Insolation / Albedo Reduction (8.5/10)

A fair amount of high pressure systems and early season open water.

4. Wind damage (6 / 10)

The Beaufort cyclone is a good example of this. Turned the gyre into rubble. Exposed a lot of surface area to ocean heat. We've also seen a tailwind pushing Laptev open water toward the pole and are now seeing Pacific water coming through the Bering Strait.

Not in the league with the 2012 GAC.

5. Export - Wind and Current (10/10)

Nares has been open. Fram has been steady. Perhaps the most important feature of the season has been the wind assisted push of ice to warm water to the Beaufort, Barents and Fram.

6. Rain (1/3)

Not much rain to speak of this year.  ;)

7. Starting Point (4.5 / 5)

2019 started the year from a low max. Less work to do.

8. Intangibles / Momentum (2/2)

Here's a category that gives the reviewer a little bit of leeway. I'll give 2019 some bonus points for strong momentum and no letup.

All in all, we come up with a score of 50.5 out of 60 possible and that is for a season that is ranking among the leaders in many categories.

This is just a straw man / food for thought. It's clear that there is a challenge in evaluating a season and benchmarking vs prior years. A system like this is intended to steer the focus toward the root causes of the end result and is easily accessible in terms of understanding.

No pride of ownership here. If someone sees a useful variant of this, feel free to run with it.

Neven

  • Administrator
  • First-year ice
  • *****
  • Posts: 7033
    • View Profile
    • Arctic Sea Ice Blog
  • Liked: 612
  • Likes Given: 407
Re: System for evaluating melting seasons ??
« Reply #1 on: June 29, 2019, 10:16:37 AM »
It's a good idea, but a lot of work, because I would suggest to do this for every month.
Il faut comparer, comparer, comparer, et cultiver notre jardin

oren

  • Young ice
  • Posts: 4205
    • View Profile
  • Liked: 688
  • Likes Given: 1193
Re: System for evaluating melting seasons ??
« Reply #2 on: June 29, 2019, 12:29:37 PM »
I would add PIOMAS volume, ice age distribution from NSIDC, Tealight's AWP, Wipneus's Inner Basin area.

kassy

  • Grease ice
  • Posts: 554
    • View Profile
  • Liked: 126
  • Likes Given: 234
Re: System for evaluating melting seasons ??
« Reply #3 on: June 29, 2019, 05:50:06 PM »
Are you sure it is 8.5/10 , 10/10? It´s hard to make each of those measures as objective as possible.

I use a crude system.

Whatever happens early on is not that important because it is the summer that decides the fate of the ice. And the seas that melt out all years don´t really matter so i watch the developments in the Central Arctic Seas where the ice is.

And in the long run i watch the freeze seasons to see which fringe areas stay open longest.

The rating that counts is the actual lowest end state of the ice which is just the beginning state for next year. 

It´s like horse racing. You want to know which horse wins, whatever they are doing by turn 2 is not important even if we get accurate mm distances.

It would be cool to have objective data on 1-6 but we don´t have those so we are stuck with SIE and SIA.
Þetta minnismerki er til vitnis um að við vitum hvað er að gerast og hvað þarf að gera. Aðeins þú veist hvort við gerðum eitthvað.

Neven

  • Administrator
  • First-year ice
  • *****
  • Posts: 7033
    • View Profile
    • Arctic Sea Ice Blog
  • Liked: 612
  • Likes Given: 407
Re: System for evaluating melting seasons ??
« Reply #4 on: June 29, 2019, 06:10:43 PM »
First you have to determine categories, like Rich has done. Then you have to assign weight to each category. And then you need a multiplier based on ranking. For instance, all the categories combined are 100 points, and you say that SLP is worth 15 points. You then compare all years since 2005 for SLP in June, that would be 15 years in total. If you think that 2019 is number 2 (by eyeballing, of course), the 15 points of the SLP category would be multiplied by 14 = 210 points.

You do that every month, and then accumulate the points. I would start in April, or maybe May.

The hardest thing is assigning weight to each category. And determining categories isn't easy, either, and they would probably change from month to month. It really would be a lot of work. I sort of do it, in an inconsistent way, for melting momentum, and still can't say anything for sure.
Il faut comparer, comparer, comparer, et cultiver notre jardin

magnamentis

  • Guest
Re: System for evaluating melting seasons ??
« Reply #5 on: June 29, 2019, 06:20:45 PM »
that's how i decided in the past which cars to purchase (when i had many LOL)

FUN aside, this is a great idea, i suggest using excel to begin with because later, onc would like to improve the table, make changes to it etc., there will always be a solid base that remains calculated and that is structured in the right order.

further i would NOT add piomas as a criteria because as much it has a value to compare PIOMAS seasons with each other because the flaws are persistent, to add obviously falsa data to a new model would make that anyways difficult task almost impossible.

all models are somehow flawed, hence one should at least reduce the know errors to the absolute possible minimum.

either way this is IMO a great idea, a bit biased probably because this is exactly who i do evaluations for many things to keep track of the various factors and being able to adapt their weight and values should need arise.
« Last Edit: June 29, 2019, 06:44:56 PM by magnamentis »

Sterks

  • Guest
Re: System for evaluating melting seasons ??
« Reply #6 on: June 29, 2019, 06:47:02 PM »
Thanks for this Rich, nice work.
To start, I would give more weight to early momentum

and would use European metrics, that is, Leo Messi for analogous (tho he's a 10 in just about anything in football/soccer)

Sterks

  • Guest
Re: System for evaluating melting seasons ??
« Reply #7 on: June 29, 2019, 08:35:40 PM »
Early NH Spring and land snow cover melt out too

be cause

  • Grease ice
  • Posts: 807
  • Citizenship .. a Lurker gets asylum
    • View Profile
  • Liked: 184
  • Likes Given: 169
Re: System for evaluating melting seasons ??
« Reply #8 on: June 29, 2019, 10:15:26 PM »
I agree with Sterks re early momentum .. I forecast a top 2 year on !st March based on my daily observations throughout the freezing season combined with Uniquorn's animation . I remember being rather excited early in 2016 that there would be open ocean at the pole . Then ,as Gerontocrat's perfectly timed graph of dispersion ( melting season post 2835) shows , whereas in 2012 ice compacted in August , in 2016 ice dispersed . A year that could have easily held the record instead becomes a footnote . And no open ocean at the pole . b.c.

 
 
2007 + 5 = 2012 + 4 = 2016 + 3 = 2019 ...

uniquorn

  • Nilas ice
  • Posts: 1272
    • View Profile
  • Liked: 583
  • Likes Given: 112
Re: System for evaluating melting seasons ??
« Reply #9 on: June 29, 2019, 11:51:19 PM »
Rich, we use a cheap low energy computer here to store all our data. It could probably calculate all the variables you consider relevant. Considering the small outlay and information input  you could offer the forum I think it's a 'no brainer' to buy one.

Rich

  • Guest
Re: System for evaluating melting seasons ??
« Reply #10 on: June 30, 2019, 01:20:29 AM »
Thanks for the feedback and positive reception folks.

I don't sense that there is a hard quantification to be achieved here by me. If someone else wants to...fine with me.

As indicated previously, the purpose of the metric is to try and ground the discussion in the fundamentals which drive the melting result. As an ice wrecking entity, the 2019 melting season appears to have no weakness and I'm trying to come up with an objective framework for assessing it. I'm all for acknowledging uncertainty  about weather to come in the remainder of the season, but the thing that we're currently watching is like a Cat 5 hurricane spinning in the ocean. It's a beast and we shouldn't be shy about acknowledging that.

Perhaps in a few weeks I'll update it and give it a pronoun (Naming suggestions welcome...hurricanes have the Dvorak Scale...) I'll give some thoughts to the suggestions posted here as well.

One last thing. I don't have the benefit of experience when it comes to prior melt seasons so my benchmarking of various attributes is more intuitive.

Once again, thanks for the encouragement.

Rich

  • Guest
Re: System for evaluating melting seasons ??
« Reply #11 on: July 02, 2019, 03:39:26 PM »
.

1. Atmospheric Heat (2019 score 8.5 / 10)

Plenty of early season heat. Recent scorching heat off Siberia. Consistency in terms of maintaining a positive anomaly throughout.

2. Ocean Heat (10 / 10)

The Pacific side is off the charts and the Atlantic is sporting a 0.5C positive anomaly.

3. Insolation / Albedo Reduction (8.5/10)

A fair amount of high pressure systems and early season open water.

4. Wind damage (6 / 10)

The Beaufort cyclone is a good example of this. Turned the gyre into rubble. Exposed a lot of surface area to ocean heat. We've also seen a tailwind pushing Laptev open water toward the pole and are now seeing Pacific water coming through the Bering Strait.

Not in the league with the 2012 GAC.

5. Export - Wind and Current (10/10)

Nares has been open. Fram has been steady. Perhaps the most important feature of the season has been the wind assisted push of ice to warm water to the Beaufort, Barents and Fram.

6. Rain (1/3)

Not much rain to speak of this year.  ;)

7. Starting Point (4.5 / 5)

2019 started the year from a low max. Less work to do.

8. Intangibles / Momentum (2/2)

Here's a category that gives the reviewer a little bit of leeway. I'll give 2019 some bonus points for strong momentum and no letup.

All in all, we come up with a score of 50.5 out of 60 possible and that is for a season that is ranking among the leaders in many categories.


I'm going for a little update and slight downgrade of 2019.

Atmospheric and ocean heat were a combined 18.5 / 20 in the initial pass. I'm downgrading here for two reasons. The intense heat near the Laptev has receded and is replaced by less acute heat in other areas. The other reason is that most of the positive heat anomaly is located in areas where the ice is already gone. That warm water / air on the Pacific side isn't optimally placed to do damage. CAA will get more action.

New score 16 / 20.

Rain

I'll bump this up to from 1/3 to 2/3. We're going to get some on the Pacific side with modest warmth and plenty of wind.

Composite score down from 50.5 to 49.

Momentum is still solid. We've lost > 2.3n km2 of area in 3 weeks and the Final Four of 2018 (CAB, CAA, ESS, and Beaufort) have lost > 100k in the last 2 days.

I'm learning as I go here. Not always pretty. I don't mind my ignorance being revealed, but important to be respectful of others. I get a zero this week. Ciao.