Sea Level and Insolation

I keep spending my climate research time involved in some ongoing discussions about some fairly specific topics.  Here is one of my replies in a discussion that involves changing albedo caused by snow in the Arctic region that becomes persistent.   The first part is my analysis of the impact of albedo on snow extent and the 2nd part is a later reply that involves sea level.




There are really two parts to the theory.

#1  Open Arctic causes greater snowfall.

#2  Greater snowfall extent increases the albedo of the Earth and increases energy reflected away.

It is possible that one could be correct, but not the other since they are not really coupled theories, but it all starts with the arctic ice.


The Inconvenient Skeptic

Seasonal change in the Arctic Sea Ice Extent


This shows the very real drop in the Arctic ice coverage.  According to the theories this should trigger an increase in global snow coverage.


The Inconvenient Skeptic

Snow Extent of the Northern Hemisphere


What is seen is different than what is predicted by a more directly open ocean.  Six months of the year are showing increased extent and 6 months are showing decreased coverage.  September-February are the months showing increased snow and interestingly it is March-August that are all showing decreased snow coverage.
I will show this another way.


The Inconvenient Skeptic

Snow extent for the Northern Hemisphere over the course of the year.


What is happening is the snow is melting faster than it has in the past, even though there is more snow falling in the winter.  However the snow extent is recovering faster in the Fall than it has in the past.

So let’s compare the expected results.  Theory #1 would predict that snow would recover sooner in the Fall and that fits what is seen, this is not proof, but it does not disprove theory #1.

Theory #2 however does not fair as well.  Higher snowfall should reflect away more energy and cause it to melt later in the year.  What is observed is faster melting in the spring which does not fit with Theory #2.  The higher albedo (associated with greater snow extent) is not causing snow to melt later and in fact the opposite is taking place.

So while the open ocean may be contributing to the earlier snow, the increased albedo is not doing anything at all as the snow is clearing faster now than it has in the past.  In fact the rate of melting is much higher now since the overall winter snow is higher than before, but it is disappearing many weeks earlier than in the past.

So my conclusion on the theories:

Theory #1:  Possible.

Theory #2:  Busted.

So I will continue to ignore snow albedo, but I will consider open arctic as a contributor to earlier snow.



Part 2 will be added shortly.




Part 2:


When I was in Houston it was mentioned to me that the open Arctic theory had the ice volume generating very quickly at the end of the Eemian interglacial. The data for the sea level that I have seen does not support this.

Siddall et al. Nature 432, 853-858 (2003)
Siddall et al. Geology 34, 817-820 (2006)
Rohling et al. Nature Geoscience 1, 38-42 (2008)

The Inconvenient Skeptic

Relative sea level for the past 140,000 years.


This shows back and forth swings in the sea level over the past 140,000 years. If the ice volume in the Arctic was sufficient to drop the sea level by 30 meters, the odds of an open Arctic are almost zero.

So from 110,000 YBP to 15,000 YBP the Arctic was completely closed in ice, but the sea level continued to fluctuate are regular intervals. This included some dramatic short term periods of warming and cooling, not full interglacial warming or glacial cooling, but in-between levels. There is no way that the changes in the frozen Arctic could have caused those swings.

The only factor I have found that could have caused those swings in the Insolation changes.


The Inconvenient Skeptic

Relative Sea Level and 65N Summer Insolation

It takes time to accumulate ice sheets as the rate of formation is limited to the evaporation rate of the oceans. Today that value is ~1m/yr. That is the maximum rate of ice sheet formation but since a majority of precipitation takes place in the Tropics, the real rate of ice volume increase is much smaller than that. Ice sheets can melt much faster than they can form.

In all cases that is observed. When the 65N insolation is low, relative sea level (RSL) drops gradually, but persistently. When 65N insolation is high the RSL can rise quite rapidly. Every time there was an increase in 65N insolation, there was a corresponding increase in RSL. ALWAYS!!!!!

Milankovitch wasn’t wrong 4 out of 5 times, it is correct in all instances, but the magnitude of the response was not always the same. That is what I spent my time trying to understand. In each case though the sea level increased by at least 30m, that is a 100 foot rise in sea level associated with every single increase in the 65N insolation.

Every drop resulted in a larger drop in sea level. Milankovitch is 5 for 5 over the past 130,000 years at predicting the direction of the change in sea level / ice volume. The timing of the swings is precise as well.

Even more impressive is the behavior from 35,000 to 60,000 YBP. Due to an unusual (400,000 year eccentricity behavior) there was a period of 25,000 years of relative stability in the 65N insolation. The overall stability of the sea level during that period is unique in the sea level data of the past 150,000 years. A 25,000 year period where the sea level stayed about 75 +/- 15m of where it is today. That is the longest such period of stability.

A sea level that was 75m lower than today would certainly not experience an open Arctic, but when the insolation dropped, so did the sea level. It dropped all the way to 120m below today’s levels. Your theory requires that the Arctic be open for that, but the stating point was an ice volume in the NH that was comparable to the Antarctic ice volume is today. And from that volume it grew even larger.


From the sea level data I can conclude several key things:

#1: The idea of a correct and stable sea level is stupid (PC version of what I really think).

#2: Insolation is the only independent factor that can explain the behavior of the sea level.
(for eaxample) Isotope ratio matches the changes in sea level, but it is changes in sea level that cause the changes in the isotope record.

#3: Looking at global changes in time periods of less than 1,000 years are a waste of time. Anything less than a 1,000 years can be termed short-term and are not indicators of long-term trends.

I play the game of #3 simply because so few people will only look at 5,000 year trends. Arguing about the behavior inside of the past 1,000 years is silly because really the only thing that matters is that the past 1,000 years is the coldest period since the Holocene interglacial began. But I am forced to play the game because people seem to care that 2012 is one of the warmest years of the past 150. Yawn!!!!!

Here is what I look at. Glacial activity over the past 8,000 years.

That paper I reviewed there shows peak glacial activity 200 years ago, minimum ~6,000 years ago. Today is so close to maximum that thinking they are going away is a cruel joke.

MWP and Roman warming periods are tiny bumps in the road. It is the trend of 1,000 year periods I pay attention to. Look at the ice volume of Norwegian glaciers from that perspective. The short-term trends mean nothing. They are real, but they predict nothing.

Posted in Snow / Snowpack by inconvenientskeptic on January 18th, 2013 at 5:42 am.


Previous Post:  

This post has 6 comments

  1. You have a classic mediating relationship here and no exogenous instrument. From the data at hand, you can’t make any inference of how the constructs are related. My sense is you are used to working with experimental data where treatments are exogenous. Not so here. Check stats and design of quasi experiments for guidance.

  2. Richard111 Jan 22nd 2013
  3. inconvenientskeptic Jan 23rd 2013


    Neither theory is mine, but one that was presented to me. This was my basic check of the presented theory. It isn’t one that I am all that impressed by, but I gave it a look.

    Since I had all the data readily available it was easy to take a look. What is interesting is that the trend in snow extent is faster melting and quicker coverage than in the past.

    I am sure the person who presented the theory to me would appreciate your comments though.

  4. inconvenientskeptic Jan 23rd 2013


    I tend to use:

    because they also have data. Granted people have been confused by the charts there when I have used them on my site.

  5. Richard111 Jan 24th 2013

    That wasn’t a recommendation John, hence the word ‘pretty’. I find people who have no understanding of AGW can look at it without going glassy eyed as they would a Rutgers chart. Handy talking point to ask “What do you think would happen if all that snow didn’t melt, or only half melted?”. Of course as soon as you mention ‘albedo’ they go glassy eyed. 🙂

  6. inconvenientskeptic Jan 25th 2013

    This whole discussion came about because there is a sub-section of people in a group I work with that is strongly convinced that snow albedo is critical.

    This was my data review of their concept. It is particularly interesting that the snow is melting more quickly in the Spring even though the albedo has been higher due to larger snow extent. I think one of the problems is that greater snow extent is that it can’t last because it is light and in regions that melt very quickly. Snow in Israel happened this winter, but it is gone quickly. So extent is high, but meaningless.

    And it really was a pretty picture. 🙂

Web Design & Dev by

Mazal Simantov Digital Creativity