Saturday 13 June 2015

Free our climate data - from Geneva to Paris

Royal Air Force- Italy, the Balkans and South-east Europe, 1942-1945. CNA1969

Neglecting to monitor the harm done to nature and the environmental impact of our decisions is only the most striking sign of a disregard for the message contained in the structures of nature itself.
Pope Francis

The 17th Congress of the World Meteorological Organization in Geneva ended today. After countless hours of discussions they managed to pass a almost completely rewritten resolution on sharing climate data in the last hour.

The glass is half full. On the one hand, the resolution clearly states the importance of sharing data. It demonstrates that it is important to help humanity cope with climate change by making it part of the global framework for climate services (GFCS), which is there to help all nations to adapt to climate change.

The resolution considers and recognises:
The fundamental importance of the free and unrestricted exchange of GFCS relevant data and products among WMO Members to facilitate the implementation of the GFCS and to enable society to manage better the risks and opportunities arising from climate variability and change, especially for those who are most vulnerable to climate-related hazards...

That increased availability of, and access to, GFCS relevant data, especially in data sparse regions, can lead to better quality and will create a greater variety of products and services...

Indeed free and unrestricted access to data can and does facilitate innovation and the discovery of new ways to use, and purposes for, the data.
On the other hand, if a country wants to it can still refuse to share the most important datasets: the historical station observations. Many datasets will be shared: Satellite data and products, ocean and cryosphere (ice) observations, measurements on the composition of the atmosphere (including aerosols). However, information on streamflow, lakes and most of the climate station data are exempt.

The resolution does urge Members to:
Strengthen their commitment to the free and unrestricted exchange of GFCS relevant data and products;

Increase the volume of GFCS relevant data and products accessible to meet the needs for implementation of the GFCS and the requirements of the GFCS partners;
But there is no requirement to do so.

The most positive development is not on paper. Data sharing may well have been the main discussion topic among the directors of the national weather services at the Congress. They got the message that many of them find this important and they are likely to prioritise data sharing in future. I am grateful to the people at the WMO Congress who made this happen, you know who you are. Some directors really wanted to have a strong resolution as justification towards their governments to open up the databases. There is already a trend towards more and more countries opening up their archives, not only of climate data, but going towards open governance. Thus I am confident that many more countries will follow this trend after this Congress.

Also good about the resolution is that WMO will start monitoring data availability and data policies. This will make visible how many countries are already taking the high road and speed up the opening of the datasets. The resolution requests WMO to:
Monitor the implementation of policies and practices of this Resolution and, if necessary, make proposals in this respect to the Eighteenth World Meteorological Congress;
In a nice twist the WMO calls the data to be shared: GFCS data. Thus basically saying, if you do not share climate data you are responsible for the national damages of climatic changes that you could have adapted to and you are responsible for the failed adaptation investments. The term "GFSC data" misses how important this data is for basic climate research. Research that is important to guide expensive political decisions on mitigation and in the end again adaptation and ever more likely geo-engineering.

If I may repeat myself, we really need all the data we can get for an accurate assessment of climatic changes, a few stations will not do:
To reduce the influence of measurement errors and non-climatic changes (inhomogeneities) on our (trend) assessments we need dense networks. These errors are detected and corrected by comparing one station to its neighbours. The closer the neighbours are, the more accurate we can assess the real climatic changes. This is especially important when it comes to changes in severe and extreme weather, where the removal of non-climatic changes is very challenging.
The problem, as so often, is mainly money. Weather services get some revenues from selling climate data. These can't be big compared to the impacts of climate change or compared to the investments needed to adapt, but relative to the budget of a weather service, especially in poorer countries, it does make a difference. At least the weather services will have to ask their governments for permission.

Thus we will probably have to up our game. The mandate of the weather services is not enough, we need to make clear to the governments of this world that sharing climate data is of huge benefit to every single country. Compared to the costs of climate change this is a no-brainer. Don Henry writes that "[The G7] also said they would continue efforts to provide US$100 billion a year by 2020 to support developing countries' own climate actions." The revenues from selling climate data are irrelevant compared to that number.

There is just a large political climate summit coming up, the COP21 in Paris in December. This week there was a preparatory meeting in Bonn to work on the text of the climate treaty. This proposal already has an optional text about climate research:
[Industrialised countries] and those Parties [nations] in a position to do so shall support the [Least Developed Countries] in the implementation of national adaptation plans and the development of additional activities under the [Least Developed Countries] work programme, including the development of institutional capacity by establishing regional institutions to respond to adaptation needs and strengthen climate-related research and systematic observation for climate data collection, archiving, analysis and modelling.
An earlier climate treaty (COP4 from 1998) already speaks about the exchange of climate data (FCCC/CP/1998/16/Add.1):
Urges Parties to undertake free and unrestricted exchange of data to meet the needs of the Convention, recognizing the various policies on data exchange of relevant international and intergovernmental organizations;
"Urges" is not enough, but that is a basis that could be reinforced. With the kind of money COP21 is dealing with it should be easy to support weather services of less wealthy countries to improve their observation systems and make the data freely available. That would be an enormous win-win situation.

To make this happen, we probably need to show that the climate science community stands behind this. We would need a group of distinguished climate scientists from as much countries as possible to support a "petition" requesting better measurements in data sparse regions and free and unrestricted data sharing.

To get heard we would probably also need to write articles for the national newspapers, to get published they would again have to be written by well-known scientists. To get attention it would also be great if many climate blogs would write about the action on the same day.

Maybe we could make this work. My impression was already that basically everyone in the climate science community finds the free exchange of climate data very important and the current situation a major impediment for better climate research. After last weeks article on data sharing the response was enormous and only positive. This may have been the first time that a blog post of mine that did not respond to something in the press got over 1000 views. It was certainly my first tweet that got over 13 thousand views and 100 retweets:

This action of my little homogenization blog was even in the top of the twitter page on the Congress of the WMO (#MeteoWorld), right next to the photo of the newly elected WMO Secretary-General Petteri Taalas.

With all this internet enthusiasm and the dedication of the people fighting for free data at the WMO and likely many more outside of the WMO, we may be able to make this work. If you would like to stay informed please fill in the form below or write to me. If enough people show interest, I feel we should try. I also do not have the time, but this is important.

Related reading

Congress of the World Meteorological Organization, free our climate data

Why raw temperatures show too little global warming

Everything you need to know about the Paris climate summit and UN talks

Bonn climate summit brings us slowly closer to a global deal by Don Henry (Public Policy Fellow, Melbourne Sustainable Society Institute at University of Melbourne) at The Conversation.

Free climate data action promoted in Italian. Thank you Sylvie Coyaud.

If my Italian is good enough, that is Google Translate, this post wants the Pope to put the sharing of climate data in his encyclical. Weather data is a common good.

* Photo at the top: By Royal Air Force official photographer [Public domain], via Wikimedia Commons


  1. As important acute it is to make climate data available -- I think it is wrong to push that into COP21. This is basically about the fate of humanity. And it hinges on agreements about mitigation.

    There is (or should) no science involved any more. We know everything that there is to know to make the crucial decisions. Distracting from the main problem by attaching our little special interests in the hope to get a lift would be wrong.

    No, better data doesn't make adaptation a replacement for mitigation. Neither the temporal nor the spatial predictability gap goes away with more historical time-series. While the topic is marginally related to adaptation capacity in the developing word, there is a long list of issues that should have priority.

    With the same arguments you are using, one could ask for more climate science professorships, larger computers, and more satellites at COP21.

    Also, you seem to think that the only stopgap for openly available climate data is the met services' desire to sell the data. From my experience, at least as important are the resources required to get that data into a publishable form, maintain updating and quality control and build the technical infrastructure to make the data available.

  2. Do you mean that when we insert a paragraph about data sharing the climate treaty will not pass because of that? If that were the case, I would also prioritize solving the climate problem, but I see no reason why such a petitesse should endanger the treaty.

    We know more than enough to get started with climate policies. I would have said that 20 years ago. That is, however, also no reason not to do more science and be able to improve our understanding and response. Adaptation will be very costly if we do not know what is coming, then you would have to adapt to every possible change.

    Adaptation is not a replacement for mitigation. Just as mitigation is no longer a replacement for adaptation. We are suffering the consequences now and it will get worse for the coming decades even with the most aggressive mitigation policies.

    I would say that research, just as weather data is a common good. In this respect it would make sense to put it in the climate treaty and as I wrote above it is already in the treaty, just like data sharing, just in a weak formulation.

    The advantage of data sharing over science funding is that it is very concrete. It is thus clear what people agree on. In case of science in general, you may be able to agree on an funding amount, but that does not automatically lead to better science, just to more spending. Like the rules of some universities that their professors get a certain amount of external funding can make scientific projects more expensive and if I were the tax payer, I would make that illegal. It would also just lead to research being called climate research that before was something else. Like development aid is to a large part export subsidies.

    To help weather services with data management, quality control and homogenization we have the Global Framework on Climate Services. That is also why I suggested to also ask for some money for the observational networks themselves. But also many well-equipped and well-staffed weather services still sit on their data. At the moment that still includes your highly advanced home country.

  3. > Do you mean that when we insert a paragraph about data sharing the climate treaty will not pass because of that?

    It just doesn't belong there. It might distract, cost time, whatever. I think it is a bad strategy to try attach any particular interest -- as important as it might be -- to anything high profile where you somehow can construct a relationship.

    There are "Open-Data" - initiatives from city- to international level, where such issues belong.

    But we completely agree on how important liberation of this data is, that it is an extremely low hanging fruit to improve research without the bad side-effects that other measures might have as you rightly point out.

    > But also many well-equipped and well-staffed weather services still sit on their data. At the moment that still includes your highly advanced home country.

    I think my home country is embarrassingly backwards in many such areas. See, that's why I emigrated ;). Look at my truly advanced country of residence: MeteoSwiss data has been always free for research and education, safe maybe a small handling fee for shady foreigners. All MeteoSwiss data will be free even for commercial use in 2016.

    The problem here (and I also believe it's the underlying problem in other countries) is inertia, the lack of technical know-how and the lack of understanding users/researchers needs inside public institutions that makes the process of actually obtaining the data very cumbersome and also expensive (for whomever). MeteoSwiss is the the most advanced agency in that regard, try downloading discharge data from BAFU ;)

    Other developed countries are making laws and embracing at least verbally "open data strategies". The real problem that needs to be pushed (and financed) in the international arena is network density, know-how, fuck EVERYTHING in the developing world. Where it really could help adaptation measures. Ever tried to get time-series from Lesotho?

  4. Try that line on my wife :)


Comments are welcome, but comments without arguments may be deleted. Please try to remain on topic. (See also moderation page.)

I read every comment before publishing it. Spam comments are useless.

This comment box can be stretched for more space.