The Greenland surface melt story of 2012

[This is a post bringing together things I have done elsewhere whilst learning how to use this platform.]

Back in July 2012 the global news media went very big on a story on the Melting of Greenland.

Greenland melt on the BBC
The original BBC Greenland Melting story

I got involved in a discussion on twitter about it, with lots of context and decided to have a go at seeing if I could put together a “storyify” on it using the contributions of a wide range of outstanding climate scientists.

That was fun, and quite popular and is here:

The Greenland surface melt story

On 25 July the media went big on a satellite measurement showing 97% of the surface of Greenland showed the signature you would expect from melt, in July. I understand there are many ways to tell this story, but this is the way I saw it develop from the perspective of a polar oceanographer.

Storified by Mark Brandon · Thu, Jul 26 2012 05:17:18

Greenland's massive ice sheet has melted over an unusually large area this summer, Nasa says, describing it as a... SCIENCE
So this is the original story posted at dawn UK time. The original press release was published 24 July 2012
Satellites See Unprecedented Greenland Ice Sheet Surface Melt - NASA ScienceSatellites See Unprecedented Greenland Ice Sheet Surface Melt July 24, 2012: For several days this month, Greenland's surface ice cover m...
So I tweeted that story - but I thought that surface melt isn't just the real story. That was the paper published in PNAS yesterday linked with this.
Greenland surface melt story: Paper in PNAS this week shows last year was heavy melt too & it did affect ice mass Brandon
That was the paper published in PNAS on Tuesday 24 July and is "Open Access". If there is ever a need for yet another "open access is brilliant" statement, it is here.
Bedrock displacements in Greenland manifest ice mass variations, climate cycles and climate changeEdited by David T. Sandwell, Scripps Institution of Oceanography, La Jolla, CA, and approved June 11, 2012 (received for review March 19,...
What this paper shows is that there is they have found a link between surface melt and bedrock displacement. This implies a link to ice mass.So bed mass displaces upwards implies that there is a reduction in ice on the land. Unfortunately I got the year wrong. It was actually a year earlier and 2010, not 2011. So I tweeted that.
doh - it was 2010 not 2011 for greenland melt - but implication is last year wld be expected increased loss too. #ineedsomecoffeeMark Brandon
But given last 2011 surface melt data:
2011 Greenland Melt Season : Image of the DayExcept for the edges, Greenland is buried in ice year round. Each winter, snowfall covers even the margin, leaving the entire island whit...
which was above the long term average (but less than 2010), we could expect a similar result to that demonstrated in the PNAS paper. So I tweeted it and made the link explicit.
Widespread surface Melting on Greenland is linked to ice mass loss & so sea level Brandon
Then I made a link to the fact that the sea ice is at a low. Sure there are weather patterns going on here but there is a background of a long term trend
put greenland surrface melt & arctic sea ice tracking at low level clear evidence of arctic amplification? Tundra nextMark Brandon
I thought about it a bit more and Jason Box ( @climate_ice ) has been working for a long time in Greenland - I have been listening to him for a long time through publications and conference presentations he has given. He published something on his blog in January 2012, which showed this was going to happen. " Blog Archive " Greenland Ice Sheet Getting DarkerFreshly fallen snow under clear skies reflects 84% (albedo= 0.84) of the sunlight falling on it (Konzelmann and Ohmura, 1995). This refle...
So the ice is getting darker, and he said "This decline is not only over the lowest elevations, but occurs high on the ice sheet where melting is limited."

If the ice is getting darker then it will absorb more solar radiation.

In June 2012 @climate_ice posted " Blog Archive " Greenland ice sheet reflectivity at record low, particularly at high elevationsAn updated compilation of NASA MODIS observations of Greenland ice surface reflectivity through 22 June, 2012 indicates that now, well in...
For me the clearest statement is:
"Perhaps most remarkable about the 2012 pattern is how much darker the snow and ice is becoming, not only at the lowest elevations around the ice sheet periphery where melting is always most intese, but in the higher elevation net snow accumulation area."

but the context is as he said "It’s not a runaway loop, just an amplifier. A record setting melt season is likely if this pattern keeps up this year." He was unsurprisingly correct. Plus there was clearly something "odd" going on with the weather.

So I tweeted that.

In Jan 2012 @climate_ice - Greenland Ice Sheet is Getting Darker in June he wrote Brandon
So the observation is what Jason Box (and I) would expect!, and something many saw coming.

But then it got more interesting. Marcus Brigstocke saw the Grauniad piece on the story.
(Marcus is one of the few standup comedians who has brave enough to takle climate change and AGW )

RT @guardianscience: Greenland ice sheet melted at unprecedented rate MB Erm... Ok? Everyone ok? With this? Ok then...Marcus Brigstocke
Dr Tamsin Edwards (Bristol University), Dr Ruth Mottram (Danish Meteorological Institute) and myself tried to help him with an explanation.
@marcusbrig Agree disconcerting... but 1 event is weather...of course if it reoccurs it's climate -- Tamsin (climate / sea level scientist)Tamsin Edwards
@flimsin @marcusbrig interestingly, 2002 had almost as wide a melt extent (briefly) but 2007, 2010 and poss 2011 bigger in total melt termsRuth Mottram
@ruth_mottram @flimsin @marcusbrig plus PNAS paper linking surface melt-> mass loss Brandon
So Marcus clearly got the message that it could not be particularly striking.
But being interested in getting it right I guess, Marcus asked if we were OK with the measurement.
@icey_mark @ruth_mottram @flimsin Scare mongering aside (as if that were possible) how concerned are you by this news? A blip or a warning?Marcus Brigstocke
@marcusbrig @icey_mark @ruth_mottram I think consensus is: watch & wait. Once is weather. But if it happens again, it's the sign of a shift.Tamsin Edwards
@marcusbrig @ruth_mottram @flimsin from me not a surprise. This in Jan 12 this is result. It supports other evidenceMark Brandon
@marcusbrig @ruth_mottram @flimsin A striking measurement that supports what could be expected - but I agree with Tamsin!Mark Brandon
@marcusbrig @ruth_mottram @flimsin I dont understand how some join dots if evidence against - but hysteria at a measurement supportingMark Brandon
@flimsin @marcusbrig @icey_mark I agree too, lots of changes measured but records too short. Makes it easier to melt faster in future thoughRuth Mottram
@flimsin @marcusbrig @icey_mark And depends on rest of summer and snow this winter (several low precip winters last few years contribute tooRuth Mottram
Leo Hickman (Guardian journalist) noted:
@flimsin @marcusbrig @icey_mark @ruth_mottram "Watch and wait" - seems to sum up climate policy, tooLeo Hickman
@leohickman @marcusbrig @icey_mark @ruth_mottram that's a different questionTamsin Edwards
@flimsin @marcusbrig @icey_mark @ruth_mottram Sure. But there does seem to be an awful lot of "watch and wait" in climate discourse :-(Leo Hickman
There's clearly a long way to go in that discussion, and by other experienced colleagues as well. I look forward to their "storifys!

But Leo then asked an awesome and obvious question. How long before we can tell whether it is a climate signal or just a weather event?

@flimsin @marcusbrig @icey_mark @ruth_mottram Serious Q, tho: how long do u wait before u lose the 'weather' caveat? Climate=30yrs, no?Leo Hickman
I tried to pull a couple more people I could think of off the top of my head who have engaged lots in the past, but also missed some excellent people who also contributed.
@leohickman @flimsin @marcusbrig @ruth_mottram I would ask @richardabetts, @dougmcneall and @markpmcc + Tamsin. They see furtherMark Brandon
Luckily some of them decided to "play" and get stuck in.
@leohickman @flimsin @marcusbrig @icey_mark 30 yrs standard, but largely due to convenience when standard agreed - there was 30 yrs of dataRuth Mottram
@leohickman @marcusbrig @icey_mark @ruth_mottram Yes, so 1 event in 30 is weather... but 5 events in 35 years might be climate.Tamsin Edwards
So Both Ruth and Tamsin are saying you need a longer time period, but it is the frequency of the "extreme" events which is important too. This is not new stuff and we have seen this sort of discussion for many climate events such as hurricanes, and droughts.

But both of them point out that it is complicated and probably worth more than 140 characters.

@leohickman @marcusbrig @icey_mark @ruth_mottram So it really depends which aspect of the system you are looking at. e.g. can't say:Tamsin Edwards
@leohickman @marcusbrig @icey_mark @ruth_mottram "extreme weather increasing": depends if extreme heat (up), cold (down), hurricanes (dunno)Tamsin Edwards
@leohickman @flimsin @marcusbrig @icey_mark prob depends on what you're measuring and if rare events, palaeo important too ask @ed_hawkinsRuth Mottram
Then ace climate modeller Doug McNeal (Climate scientist and statistician at the Met Office Hadley Centre) said.
@leohickman @flimsin @marcusbrig @icey_mark @ruth_mottram This is an *interesting* question...*system running*... a) it depends.Doug McNeall
He is not being difficult here - he is being good. You really have to define the question carefully in the Earth system. At one level it is trivial. Increase the temp of the planet by say 10C, there will prob be no ice on greenland in 5000 years. But what we are asking is about rates of change. This really is hard - but vital to understand if you want to plan for example sea level defences.

Leo (quite rightly) asked about a specific guideline for the date we can expect to "know".

@flimsin @marcusbrig @icey_mark @ruth_mottram Thnx. So when is Year 1 in this case (Greenland)? When will the 30 years be up so we can know?Leo Hickman
I pointed out that this is probably not year one,
@leohickman @flimsin @marcusbrig @ruth_mottram It's not year one. As ruth said the trend is there.....Mark Brandon
SVS Animation 3720 - Annual Gradient Melt over Greenland 1979 Through 2009NASA GSFC Scientific Visualization Studio
And remember Ruth had already said (repeated here)
@flimsin @marcusbrig interestingly, 2002 had almost as wide a melt extent (briefly) but 2007, 2010 and poss 2011 bigger in total melt termsRuth Mottram
So it is almost certain this is not year "1".

Given the climate signals in the rest of the Arctic e.g the clear downward trend of Arctic sea ice extent, e.g

or a slightly longer term view

The fact that "The Arctic is warming two to four times faster than the global average"

And so on.

The question is "how much are humans responsible for". This is a serious mathematical challenge but it can and is being studied with some very interesting success. Mark McCarthy (UK Met Office) said:

@icey_mark @ruth_mottram @dougmcneall @leohickman @flimsin @marcusbrig we get around weath-climate issue with fraction attributable risk estMark McCarthy
What he means here is you can estimate the fraction of the melt attributable to humans - and it has been done here:
@leohickman @icey_mark @ruth_mottram @dougmcneall @flimsin @marcusbrig some context and examples here.. McCarthy
It's a heavy read for the non-scientist but it shows ways you can estimate what "we" are responsible for. Unfortunately to Doug's knowledge no-one is doing it for Greenland yet. But it maybe someone is.
@leohickman @icey_mark @ruth_mottram @dougmcneall @flimsin @marcusbrig last tweet example of method but has not been done for Greenland ice.Mark McCarthy
Ed Hawkins (Climate scientist at University of Reading) runs an excellent www site called the Climate Lab Book. (He has been encouraging other scientists to contribute to it).
Climate Lab Book | Experimenting in open climate science - open for contributionsThe time at which the signal of climate change emerges from the 'noise' of natural climate variability (Time of Emergence, ToE) is a key ...
Well Ed pointed out that the more variable a signal is, then the longer it takes for a signal to emerge.
@leohickman @ruth_mottram Depends on amplitude of variability which determines when 'signal' emerges from 'noise'. e.g. Hawkins
What he means is: if the temperature of London, UK in December is consistently 10-11C for 30 years and then for the last 2 years it jumps to 15-16C you can be pretty sure the signal has changed. (This is not the same as who is responsible for the signal change!).
But if the temperature of London, UK in December is consistently 10-15C for the last 30 years, and then for the last 2 years it changes to 11-16C, then you have a harder time suggesting the signal has changed.
Broadly speaking this is exactly what Tamsin, Ruth and I were saying to Marcus above.

Doug put that explicitly.

@leohickman So the number of years needed to determine 'climate' depends on which part of that distribution you are looking at.Doug McNeall
But the point remains: It can be done. Gareth Jones (Who I think is also at the UKMO but no doubt he will correct me!) gave another example in the scientific literature.
@leohickman @flimsin @marcusbrig @icey_mark @ruth_mottram This study discusses the challenges of climate attribution S Jones
John Kennedy also pointed at a world meteorological organization paper which describes the method.
@dougmcneall @leohickman Blair Trewin wrote WMO note on climate normals. It depends on the variable & intended purpose. Kennedy
And Today 26 July there is an brilliant example by  @jonny_day which is published in Environmental Research Letters has a striking result.
Loss of Arctic sea ice ‘70% man-made’ @jonny_day, formerly at Bristol. See also by @jamesannanTamsin Edwards
This is an open access paper too and worth visiting for the video abstract. Nice work Jonny.
Sources of multi-decadal variability in Arctic sea ice extent - Abstract - Environmental Research Letters - IOPscienceThe observed dramatic decrease in September sea ice extent (SIE) has been widely discussed in the scientific literature. Though there is ...
I leave the last word to Ed
@leohickman @dougmcneall Ask N climate scientists a 'simple' question. Get N different less simple answers!Ed Hawkins
I don't take this in a negative way - what he is saying (IMO) is that tying down the question is very important.

Thanks to all the contributers.

(or it doesn't matter how hard you try....)
Terrifying: massive ice melt across Greenland. Worse & faster than anyone imagined. #climatechangeGeorgeMonbiot
This is what climate change looks likeAn epic struggle for the future of our world is unfolding in the Arctic, as oil giant Shell is ready to swoop and begin drilling the land...
And inevitably that is the fault of the scientists.

Bad bad media people who use scientists!

the 'headline' have gone around the world.. job done.. scientists used. @ed_hawkins @icey_markBarry Woods
But finally I would like to make the point that press & comms officers who work in the organisations I have worked in and had dealings with have IMO been both brilliant and constructive.

The fact is they don't want to get it wrong either.


The reaction was surprising and very positive. The science story ended up with almost 3,000 views (so far) and was described by probably my favourite science journalist as “bloody excellent“.

My conclusion? You can tell complex scientific and environmental stories using storify.