Sharing data on agricultural trials: towards a global repository
We’re here in Nairobi, Kenya, in the “Multi-site trial database for climate change analysis: Planning and launch Workshop”. This is an activity being developed by Theme 5 of the Challenge Program on Climate Change, Agriculture and Food Security, and we have BIG ambitions!
We’re coining the phrase Open Access Agriculture, and the philosophy is that of making agricultural research data available. We’re starting with data on agricultural trials, whereby new technologies are tested in the field in specific sites. This is standard practice in agricultural research, but unfortunately what is not standard practice is the sharing of that data. This is essential for climate change adaptation – we need to know the limits of abiotic adaptation for different technologies. Our aim is to build on the very successful Africa Trial Sites project by populating the website with actual performance data. This will help agricultural development today, but also provide essential data for us to further understand the impacts of climate change and develop strategies for adapting into the future. Just recently Jeffrey Sachs published an article in Nature calling for a monitoring system. We’re one step ahead of him.
Some of the potential uses and benefits from having this data:
- Data for calibrating and validating crop models for understanding yield today and into the future
- Develop and inventory of promising agricultural technologies that are under development or released (a la Sachs et al.)
- Provides the data for designing “holistic” adaptation packages for specific sites and regions (for example, combining the best crop variety with the best soil management technologies and with the most appropriate water management)
- Opportunity for performing untapped GxE analyses
- A whole host of unknown or unexpected uses which we’re eager to see happen (if you build it, they will come)
Today we discussed a range of issues relating to the constraints and solutions to people actual sharing this data, and began to look into the kinds of data we need to document and share. The good news is that there is plenty of acknowledgement that this data has value and should be open access. The difficulty is in managing these complex datasets in an effective and pragmatic way. Tomorrow we go into more detail on this, and start thinking about data sharing agreements. Will provide more updates tomorrow, but things are looking good based on today’s discussions.