Nothing like a balmy summer afternoon.
Earlier this year scientists were given a survey on their opinion of the state of climate science. This was administered by Dennis Bray & Hans von Storch at the Helmholtz Zentrum Geesthacht. (Bray I don’t know, but von Storch I do, vaguely).
The paper is on-line here (free registration required). Update Alternate link at Bray’s site (near the top).
The authors started with 5,947 (reasonably discovered, mostly senior folks from USA, Germany, and UK) email addresses (culled from earlier surveys), but had to toss 1,456 for invalidity. Only 286 people turned in a survey. I was one. This makes for a very dismal 7% response rate. Any conclusions drawn from this study should therefore be viewed with fish eyes, because 93% had noting to say, did not to participate, who knows why.
What follows is a summary. Most questions were on a 7-point scale, higher more confident, increased significantly, that sort of thing. I dichotomize these, with 4 (neutral) and above or 3 and lower.
Main: No Consensus
Only 8% (of the 7%) said their “confidence in the findings of climate science” decreased. Which is to say—and not for the last time—there is no consensus. Of the other 93% who didn’t turn in a survey, nobody knows. But there are at least some who aren’t so happy with the state of affairs in climatology (I’m one).
On a bright note (to me), about 36% did not agree that “climate science has remained a value-neutral science.” But no consensus.
Around 11% felt “less confident concerning the IPCC’s attribution of warming to GHS”. No consensus in the 7%.
Model Modules: No Consensus
Some 20% did not agree that “Climate models accurately simulate the climatic conditions for which they are calibrated.” No C. Same number of folks disagreed that atmospheric models deal well with hydrodynamics. Only 10% were skeptical of modeling radiation. But 26% worried about simulate vapour in the atmosphere.
And a whole 60% admitted that climate models don’t do well with “the influence of clouds.” About half had the same negative view of precipitation, and between 50-60% frowned on atmospheric convection. Gee, No C.
The same pattern repeated itself for ocean modeling, so I won’t repeat it, except to note that 24% did not think models had the “ability to couple atmospheric and ocean models.” No C again.
What about turbulence in climate models? Just under half said no confidence; 28% said nope to land surface processes; about the same were dim on sea ice. Least negative were views on surface albedo and “green house gases emitted from anthropogenic sources”; about 14% were negative on each. No C.
Model Mimicking: No Consensus
About 9% were skeptical that models were able to reproduce both “mean values for the last 50 years” and “trends for the last 50 years”. More than double that (around 21%) were skeptical about reproducing “variability for the last 50 years.” Some 24% didn’t think models did well with precipitation over the last 50 years.
Even more—37%—said the models could not reproduce “trends for the last 50 years.” And even more still (52%!) thought models blew it on “variability for the last 50 years”. Talk about no consensus! (Of the 7%.)
Model Predictions: No Consensus
Despite that half thought models stank at reproducing variability, only 25% (why not the same 50%?) or so thought models would not well predict “mean values for the next 10 years” nor would they well predict “trends for the next 10 years.” And even more, around 38%, didn’t think models will do well with “variability for the next 10 years.” No C again.
It was the same story for predicting 50 years ahead, with even more folks skeptical of models making such long-range predictions.
Predicted precipitation? 42% said no to “mean values for the next 10 years”; 54% or so said no to “trends for the next 10 years”; and around 68% said no to “variability for the next 10 years.” Out 50 years, and skepticism only grows (as it should). No C.
Sea-level rise? Around 19% said no to good predictions of “mean values for the next 10 years”. About 23% said no to “trends for the next 10 years”, and about 31% said no to “variability for the next 10 years.” As before, out 50 years and only a handful believe. No C.
Extreme events? Over half (52%) didn’t think models would do well predicting “mean values for the next 10 years.” Around 60% said no to “trends for the next 10 years” and some 66% said no to “variability for the next 10 years.” Once more, looking out 50 years produces very little confidence. No C.
The authors also asked a series of questions on regional models, which produced less agreement than the global models; indeed, the majority were skeptical on many questions. No Consensus discovered. (Of the few who bothered to answer.)
Impact!: No Consensus
Around 28% didn’t think they had much to say about “the detrimental effects that climate change will have on society”.
Pay attention: The closest to a Consensus was to the question “How convinced are you that climate change, whether natural or anthropogenic, is occurring now?” Only 2% disagreed. Now, if even this banal, harmless question (“natural or anthropogenic“) cannot produce a Consensus, then what can?
About 11% were not convinced “that most of recent or near future climate change is, or will be, a result of anthropogenic causes”; about 14% were not convinced “that climate change poses a very serious and dangerous threat to humanity”. Note the words very serious.
Some 9% didn’t think we were feeling effects of a changed climate yet. But 38% said we could not “ttribute recent climate related disasters to climate change (anthropogenic or otherwise)”. Note that “anthropogenic or otherwise”.
The hardest question to summarize was this: “Since 1850, it is estimated that the world has warmed by 0.7 degrees C. Approximately what percent would you attribute to human causes?” About 10% said thirty-percent attribution or less. The peak was 26% at eighty-percent attribution. But, no consensus.
View: No Consensus
There then followed a series of questions on what interactions climate scientists had with the public. But about 60% of scientists said that adaptation is better than mitigation when dealing with climate change problems. Whoa!
Again, around 60% disagreed with the practice which some scientists employ; those who “present extreme accounts of catastrophic impacts related to climate change in a popular format with the claim that it is their task to alert the public.” (Bad news, right, Gav?) But still no consensus.
Even 33% said it was not the job of climate scientists to “be directly involved in alerting the general public about the possible socio-economic consequences to humans (health, policies, damages, economic loss, etc.) resulting from changes in the climate”
There then followed a few more questions along the same lines, all pointing to mixed views on the proper role of scientists and public policy.
Indulge Me: You can skip this section
The last few questions were of interest to statisticians. To “A description of the most probable outcome best defines” 28% said “a projection”, 64% said “a prediction”, and the rest “other.”
To “A description of a possible outcome best defines” 62% said “a projection”, 16% said “a prediction”, and the rest “other.”
To “From a scenario simulation prepared with climate models, scientists are more inclined to make” 76% said “a projection” and 18% “a prediction” and the rest “other.”
Now, since according to logic, all predictions are conditional (as all probability is condition), there is no difference in “a projection” or “a prediction.” Though I have the idea more people would like to hide behind the former, as it sounds weaker. More on this another day.
I’m struggling to tie a theme together. Maybe readers can help me?