24022010

Met Office proposes to make Data Set Verifiable

As reported in CA and FoxNews, with the proposal here it looks like the Met Office wants to make a central repository of climate data available to all. The real question to me is this actually going to improve the end quality of the research being done?

The climatic scientists have had a very hard time to date maintaining clean data and providing the level of quality in record keeping required for such key data - so having a special body in charge of just the data is step in the right direction. My real concern is that we do not end up with the GIGO (Garbage In Garbage Out) problem. In other words they create a perfect framework for climate data infested with 'bad' data.

To me what needs to be done is that each source of climatic data can be tracked back to it source, and that any error rates or assumptions are similarly traceable. In this way it will be very hard indeed to insert bad data into the central store as it won't have the 'credentials' required to enter.

Also, as I have mentioned previously if the data is public, then any adjustments which are made to the data also need to be made fully publicly available. Then and only then will it be possible for interested people to fully verify the data.