PLEA Network

10 February 2017, Carbon Brief, Guest post: Why NOAA updates its sea surface temperature record. The National Oceanic and Atmospheric Administration (NOAA) is one of a number of climate agencies that pieces together global temperature from thousands of measurements taken each year across the world’s land and oceans. Last weekend, an article in the Mail on Sunday sparked interest in the way NOAA constructs its temperature record. The claims in the article, widely rebutted shortly after its publication, focused on the most recent version of NOAA’s sea surface temperature (SST) record. I have been involved in the development of this dataset since 2011 and it is due to be updated again shortly. However, an early draft of the journal paper about this update appears to have been circulated amongst media without the permission of the authors (including myself). I have, therefore, decided to make some observations here in a personal capacity that may help make better sense of how the dataset is produced and what it shows. Peer review The status of the new version of our dataset – nominally labelled “ERSSTv5” – is that we have submitted a paper to a journal, and it is undergoing peer review. As is the academic norm, the authors wish to respect the review process and did not give their permission for the draft to be shared with the media. To be clear, the authors have expressly not given permission for the draft to be used or quoted in the media. To be equally clear, at this juncture, copyright of the paper remains with the authors. The peer review process on the paper has only just begun, therefore it is premature and incorrect to analyse the dataset in detail before the paper is published. Peer review is an essential step towards eventual acceptance of any new research. It is highly unusual – and, in my view, undesirable – to discuss the specifics of submitted manuscripts in a public manner before this process has concluded. Peer review will, inevitably, point out ideas which shall serve to strengthen any given analysis. I shall, therefore, not be discussing specific scientific aspects of the draft paper and shall refer to it below only to the barest extent required. If and when the paper is published, this will be the appropriate time to discuss its findings in depth and I would be delighted to do so. Read More here