Surfacetemperatures.org

Readers may recall some months ago that the Met Office planned to “do over” the surface temperature data sets:

Met office pushes a surface temperature data “do over”

The effort has started, and a website has been setup at http://www.surfacetemperatures.org

They write:

Surface temperature datasets for the 21st Century

To meet 21st Century requirements it is necessary to reconsider our analyses of historical land surface temperature changes. This is about much more than simply re-engineering existing datasets. These datasets were adequate for assessing whether climate was changing at the global scale. This current exercise should not be interpreted as a fundamental questioning of these previous efforts. But these pre-existing datasets cannot answer all the questions that society is now quite rightly asking. They do not constitute a sufficiently large sample to truly understand our uncertainty at regional scales. At monthly resolution they are also of limited utility in characterising extremes in climate and their changes.

Regardless of the causes, climate is not and never has been stable. Changes in climate impact all of society. But it is not changes in the monthly mean at scales of hundreds of Km that impact us all. It is changes at much more local scales and that last a few hours or days that have a major impact upon society. Whilst the long-term changes characterised by the current datasets can ameliorate or exacerbate the effects these datasets are fundamentally ill-suited to meet most of our needs as a global society.

The challenge set out in the proposal from the Met Office to the Commission for Climatology (CCl, see background link) is to produce a new suite of datasets capable of answering these questions. This requires more than the work of a single institution to do at all. It certainly requires very many partners to do properly. Following the positive outcome of the CCl deliberations an organising committee has been convened and the Met Office will be hosting a workshop in September in Exeter, UK. The organising committee, with substantial international representation, includes a broad range of expertise and perspectives and is undertaking planning activities. As yet planning is at too early a stage to publish details. More specific information will be posted by mid-June. However, the expected outcome of this meeting will be an in-depth plan with multi-institution sign on as to how to proceed. Broad aspects to be covered will be:

• Data recovery, digitisation and provision;
• data homogenisation and homogenisation system performance benchmarking;
• and communication, engagement and auditability

The plan is to solicit white papers on the range of topics to be discussed in advance of the meeting and post these on a (moderated) blog for broad input so that non-participants in the meeting can still have some meaningful input. The invitee list includes representatives from a number of relevant disciplines including a number that need to be engaged if the project is to be a success: climatologists; metrologists (measurement scientists); and statisticians amongst others. To be effective the meeting will have to be relatively small but, as stated above, stringent efforts will be made to entrain input from non-attendees in advance. And, of course, participation in the work will not be limited to attendees of this initial planning meeting only – to be successful it needs lots of participants, many more than will be at the meeting.

Update: 7/26 A revised version of the agenda is now available with only minor changes from the original. The white papers have begun to be posted at http://sites.google.com/a/surfacetemperatures.org/home/whitepapers (all hopefully to be posted within a week) and a moderated blog for public comments is available from http://surfacetemperatures.blogspot.com/ until August 23rd.

Update: 8/20 New version of agenda and comment period will remain open until September 1st.

==============================================

Here’s my take on it:

1) The effort, while noble, is a reaction to a series of data transparency blunders rather than a proactive approach to open replication. In the original Fox News article I cited on Feb 23rd, 2010 they write:

At a meeting on Monday of about 150 climate scientists, representatives of Britain’s weather office quietly proposed that the world’s climatologists start all over again to produce a new trove of global temperature data that is open to public scrutiny and “rigorous” peer review.

After the firestorm of criticism called Climate-gate, the British government’s official Meteorological Office apparently has decided to wave a white flag and surrender.

While this effort is a step forward, it is unfortunate that it took Climategate to break free the idea of open and transparent data, and of surface data that has gone through rigorous quality control procedures. As we’ve seen recently, Canada’s own surface weather data is in such a mess that Environment Canada squelched their own embarrassing internal report and it took a freedom of information request to pry it loose. They called the state of the network “disturbing’.

2) This statement from the white paper 3, item 8 discussed here says:

A parallel effort as an integral part of establishing the databank is required to create an adjunct metadata databank that as comprehensively as feasible describes known changes in instrumentation, observing practices and siting at each site over time. This may include photographic evidence, digital images and archive materials but the essential elements should be in machine-readable form.

Is essentially a stamp of approval of my surfacestations.org project. Without knowing the changes in measurement conditions surrounding the century long experiment in climate monitoring, it is impossible to know the true quality of the data. I see this as a positive step forward.

3) Making this effort known to the climate community has apparently not been a strong suite of the Met Office, for example, I only found out about it a couple of days ago via a reporter asking questions about my views on it.

The Met office needs to be far more proactive in communications.

I encourage readers to make submissions before the Sept 1 deadline, as only a few days remain.