Climate scientists specializing in future sea ice predictions made some remarkable statements to polar bear scientists at their last meeting – admissions that may really surprise you.
Back on June 26 (reported here), the IUCN Polar Bear Specialist Group (PBSG) posted a summary of its last meeting. So, I was very surprised to find (while there looking for something else), that on 18 July 2014 they had added minutes from the meeting to that summary.
These minutes are a bonanza because among the juicy nuggets of information is a summary of what the three invited climate scientists from Colorado (Jennifer Kay, Mark Serreze, and Marika Holland) had to say and what questions were asked. While real transparency would have involved posting copies of the sea ice presentations and transcripts of the question and answer sessions, this is certainly better than nothing.
I’ve pulled some quotes from the minutes on the presentations made by sea ice experts J. Kay (National Center for Atmospheric Research), M. Serreze, (National Snow and Ice Data Center), and M. Holland (National Center for Atmospheric Research). Some of their most illuminating remarks were answers to questions posed by PBSG biologists.
Keep in mind that the quotes below do not necessarily represent exactly what was said (i.e., word for word). They do, however, represent what the polar bear scientists keeping notes1 heard, which is probably what’s important. See the entire document here for the context of these quotes. [I’ll report in a couple of days on some of the other content]
Bold in the quotes below are mine; I’ve added notes in square brackets after some but have left the rest to speak for themselves.
“In terms of understanding what controls the sea ice extent, climate models can help us understand the year to year variability (e.g. increase in sea ice in 2013).” Pg. 10 [said J. Kay]
“S. Belikov [PBSG, Russia] then asked why no models predicted the 2012 sea ice cover. M. Serreze replied that models are not built to predict a given year because they are parameterized for the mean conditions. He emphasized that the statistics of the weather in these models are correct but that the sequencing of events are not expected to line up with reality.” Pg. 10 [This appears to have been accepted as an adequate answer]
Kay noted that model predictions indicate that sea ice has the potential to temporarily expand in a warming world over 10 year time scales (i.e., in any one decade you could see a relative increase in sea ice cover); however, projections of 20 years tend toward sea ice decline.” Pg. 10 [So, we can expect sea ice to increase over the next 10 because of global warming but if we wait 20 years, we will see that it has sort of declined]
“J. Kay concluded…that 1) summary observed trends in 1979-2013 sea ice cannot be explained by natural variability; 2) Arctic sea ice extent could increase or decrease over the next decade and that all future conditions are highly dependent on model physics, and 3) large ensembles from credible climate models are needed to understand Arctic sea ice trends in a warming world.” Pg. 10 [Got it: the models can’t predict how variable the ice may be over the next 20 years but the experts know for sure that the recent changes are not the result of natural variation]
“There were questions regarding what the summer ice might look like mid-century and whether it would be ice free. M. Serreze indicated that the range at which you reach an ice free Arctic will likely happen over a 20 year period and that there was no exact date because there are a range of model predictions; however, generally the models tend to predict an ice free Arctic some time mid-century. Pg. 11 [If sea ice models are only correct to within some “20 year period,” how does that imprecision impact the requirement of conservation status definitions that polar bear numbers must be predicted to decline by at least 30% over 3 generations (30-45 years, depending on how its calculated)?]
“He [Serreze] discussed whether Arctic amplification leads to “whacky” weather or what is referred to as stuck weather patterns and explained how the polar vortex is related to climate conditions in the Arctic….Mark indicated that there was lots of controversy about this notion but that there is model evidence that reduced sea ice cover is likely to influence weather in mid-latitudes.” Pg. 11 [“model evidence” – what’s that? Wouldn’t “model output” be the correct term?]
“Were AO and NAO variability related to sea ice loss? M. Serreze noted that these indices can be very limited because they are influenced by certain conditions in certain places.” Pg. 11 [Is this supposed to be an answer?]
“Polar bears have seen warming climatic conditions before – are there good estimates of what sea ice looked like in the paleoclimate record? M. Serreze indicated that there is not a lot of data and that it is hard to reproduce the historical sea ice record.” Pg. 12 [They can’t “reproduce” the historical sea ice record exactly, so nothing about past sea ice extent is relevant to the polar bear’s response to sea ice changes?]
The PBSG biologists asked questions that sound suspiciously like the kind of queries they get from the media or audience members when they give public lectures. As a consequence, they seemed more interested in having pat answers to pass along than in demanding clarification on nonsensical or troubling responses from the ice experts. Was that the purpose of these presentations?
Was no one concerned with the admission that ice models could not give a prediction more accurate than plus/minus 20 years? Twenty years represents about half of their required window for a predicted population decline – surely that will affect the validity of any model trying to predict how polar bears might respond to projected sea ice changes over the next 30-50 years. But if any of the PBSG biologists noticed, they either didn’t say or it didn’t get included in the minutes.
Was no one concerned that the sea ice experts could not reproduce the historical sea ice record but were absolutely confident that the changes seen between 1979 and 2013 could not be explained by natural variation? If so, it is not reflected in this document.
I wonder — do we really need to “reproduce the historical sea ice record” to understand the significance of the fact that polar bears suffered only a modest population decline during the last major interglacial, when sea ice was as low as has been predicted by ice models for the end of this century?
We do, in fact, have data on the lowest extent of sea ice reached during the Eemian and that’s what really matters: ~115,000-130,000 years ago, the Arctic was virtually ice-free in summer and there was no ice in the Bering Sea in winter. And yet, polar bears survived without getting anywhere close to extinction.
Here’s the questions that occurs to me after reading this document: is the conservation focus of PBSG biologists keeping them from being good scientists?
Footnote 1: Note-keepers of the minutes (“meeting rapporteurs”) were listed as Steve Amstrup, Todd Atwood, Dena Cator (IUCN rep), Andy Derocher, Kristin Laidre, Nick Lunn (lead), Evan Richardson, and Geoff York (PBSG 17 Minutes: 1).