A draft of the United Nations organisation's fifth climate report (IPCC AR5), due to be completed 2014, has been leaked onto the internet.
The International Panel of Climate Change is a time-consuming voluntary process comprising three working groups, that produce the three blockbuster reports (on physics, impacts, and mitigation strategies respectively) every seven years. Although the process produces no new science, it is intended to faithfully reflect the state of play.
Although anyone could sign up to "review" the draft, anyone who did was forbidden from disclosing the contents. Skeptical climate blogger Alec Rawls signed up and waited until the after the second review round closed on November 30th, then posted the draft to his blog - which promptly buckled under the weight.
Rawls' decision has won some support from surprising quarters. The New York Times Malthusian-in-residence Andy Revkin described the IPCC demands for secrecy as archaic.
What does it say, then? We shall refrain from cherry-picking all but the cherry-picked "Summary for Policy Makers". There's little change in the conclusion since AR4. Although only a small fraction of CO2 in the atmosphere is produced by human activity, and only half of that isn't absorbed, this remaining component is the dominant factor in determining the climate, say IPCC authors.
"Globally, CO2 is the strongest driver of climate change compared to other changes in the atmospheric composition, and changes in surface conditions. Its relative contribution has further increased since the 1980s and by far outweighs the contributions from natural drivers."
Computer models predict a 3°C rise over a hundred years, and they're more sure than ever "that many changes, that are observed consistently across components of the climate system, are significant, unusual or unprecedented on time scales of decades to many hundreds of thousands of years."
Dire predictions of an increase in violent weather aren't justified however, according to the IPCC.
Rawls says he was motivated to leak the draft because of the IPCC's refusal to accept any other solar factors other than irradiance, or 'brightness'. The refusal is a fact, but results largely from the failure of the scientific community to explore such factors. The Svensmark hypothesis on cloud nucleation is at least acknowledged, for the first time.
It is rarely noted that the IPCC process is voluntary, and the more distinguished skeptical scientists - those who accept the CO2 hypothesis but attribute lower postive feedbacks for CO2 in the climate model, such as Professors Christy and Lindzen, dropped out after AR3 was published in 2001. What's left is scientists with the time to participate and Governments (or more accurately, state apparatus: the bureuacracies) with the inclination and funding to take part.
If Revkin described the confidentiality demands as archaic, much could be said of the entire process, which is designed to chivvy policy-makers into adopting carbon mitigation policies (aka into reducing CO2 emissions), as a sister process to the United Nations Framework Convention on Climate Change talks. Yet it is now evident, after the failure of successive conferences, that most countries favour adaptation rather than mitigation. China has surpassed the USA in CO2 emissions and now adds an entire "USA's worth" of CO2 to the atmosphere on top of that. This leaves a rump of Anglophone countries (the UK, Australia and New Zealand) still pushing for mitigation - allied to a cluster of small states looking for as much compensation as they can extract from the process. Thus the IPCC appears to be designed for a bygone age.
You can find links to the leaks to the AR5 WG1 draft all over the place, including via the Watts Up With That Blog here. ®
The Interacademy Panel made several recommendations to improve public confidence in the IPCC (pdf), after much criticism.
The WG3 part of the IPCC process was criticised for its heavy reliance on "grey literature" and the role of environmental activists in the editing process). It is left as an exercise for the reader to decide how much was incorporated into AR5.