May was all about Agroknow’s Open Harvest! The logic behind this meeting has been explained in detail here and one of the major outcomes is the -already famous- Chania Declaration. As promised and planned, on Thursday 19th of May, second day of the meeting and after the end of all discussions and presentations, it was time to offer all participants a first sneak preview of the use case pilot demonstrator that we have been developing in the context of the BigDataEurope Horizon 2020 project. In order to help participants visualize what this demonstrator will be all about and what problems could help solve, we had organized a hands-on-visit to a nearby traditional vineyard and winery. We were welcomed by Antonis of the Dourakis Winery who was kind enough to take us for a tour in the family’s vineyards and present the current status of Greek viticulture from the commercial point of view. At the same time, and amidst our in situ presence in the vineyard, it was the perfect opportunity to connect this commercial perspective with the researcher perspective and explain what the grapevine varieties research activities are. A key point is to see how all this interconnects and how problems that are solved through research and applications like our use case pilot could have direct correlation and impact on wine growers and wine makers. After the visit in the vineyard, we had a quick look at the winery and then it was time for the Vitis Big Data Demonstrator (SC2 use case) presentation, accompanied with a fine winetasting. Participants had the opportunity to taste some of Crete’s local wines from indigenous grapevine varieties while we presented them with the pilot demonstrator and examine what the challenges on typical data collection, management, processing and visualization are. Feedback was collected through discussions on the pilot from different perspectives. Some of its components and the logic behind the demonstrator can be found here and here, while more details and interesting things will follow as the development continues. Stay tuned!
In an earlier post, we explained how Variety, one of the four Vs of Big Data, applies in a specific domain as is the one of Viticulture and how all this connects with Agroknow’s work (a.k.a. our use case pilot) in the context of the BigDataEurope Horizon 2020 project.
Viticulture is one of those domains of Agriculture, and obviously not the only one, where data are present throughout the entire process, from the vineyard where various activities take place to the laboratory where analyses are being done. Different vineyards to set the experiment, different grapevine varieties or even different samples of the same variety but from different locations to study, different equipment to use, different methods and protocols to follow, all these contain more than one type of data and all these constitute variables that can differentiate the research outcomes.
One of the challenges that we wanted to tackle with our use case pilot is to see how difficult it is to gather all this information and at the same time, to find ways in which we can organize and combine all these heterogeneous data produced in order to visualize this information and reach meaningful conclusions.
For example, before doing anything else, a viticulture researcher has to decide and locate which varieties to study for the purpose of an experiment. A usual procedure for our researcher is to visit a vineyard, choose a variety that he/she wishes to study and then select a specific plant-sample of the chosen variety that meets certain criteria of appearance, health etc. This chosen plant has specific geolocation or geotagging so that we can revisit it at any time, and also it comes with a set of data. The researcher will have to: take a picture of this plant, take samples from various parts of the plant for specific measurements (young shoot sample, leaf sample, bunch sample with the corresponding pictures) and analyses (genetic study), and the researcher will have to do this during different time periods throughout the year (depending on the different stages of plant growth) and for a number of consecutive years.
Eventually, our researcher ends up with a great number of different spreadsheets, different pictures and different analyses and measurements that are all related with a vineyard, a variety and a plant-sample. And what happens when our researcher needs to do the same thing for many more varieties and for many more plant-samples? How can big data (analytics, visualizations and other tools) can help organize and combine all this information?
Wouldn’t it be exciting but above all totally useful to be able to access or view all these different and various data related to the sampling process and have them visualized with just a couple of clicks? Let’s wait and see what our use case pilot has to say about this.
The Wageningen University and Research Centre (WUR) is a Dutch public university in Wageningen, Netherlands, which consists of the Wageningen University and the former agricultural research institutes (Dienst Landbouwkundig Onderzoek – DLO). WUR is one of the top institutes at a global level in the field of agri-food and environmental research. As expected, such a large and active group of research institutes produces huge amounts of data – and WUR has developed the expertise throughout the years to make good use of big data in the agri-food research context. It has been almost one century (98 years, to be more precise) since WUR started collecting research data of various types, using various means and managing all this information and data so that it can be easily reused. How has this data been collected through these years, for which purpose and how can it benefit future research? A beautiful timeline produced by WUR provides the answers to these questions.