The Need for Non-Conventional Data Sources to Improve the Sustainability of Agri-Food Systems in the Developing World
This blogpost is written by Dharani Dhar Burra, who is a visiting scientist from CGIAR Platform for Big Data in Agriculture. Dharani will be with us at Pulse Lab Jakarta on secondment for the next six months working on projects that will use non-conventional data sources to better understand, and derive new insights about the complex relationships between agriculture production, their impacts on environment and food consumption.
Agriculture has been and is today still the mainstay and backbone of several developing countries’ economy. In these countries, a large proportion of the population is both directly and indirectly linked to agriculture. Recent trends related to economic activities in these countries show that other sectors, such as the service and manufacturing industries, have seen significant improvement, and are pulling in more workforce. The younger generation is gradually transitioning to these sectors to ensure more sustained and even higher incomes. This is in stark contrast to agriculture, wherein climate variability and economic shocks have increased uncertainty in incomes.
Improved incomes of population in general have several manifestations. An interesting trend to watch out for is food purchase and consumption patterns, that have changed significantly over time. It is important to assess these trends in the backdrop of changing agricultural production landscape, of jobs in the alternative sector with higher incomes and higher uncertainty in agricultural productivity, and many other drivers such as globalization.
Agri-food systems are defined using the concept of farm to fork, or production to consumption etc., and all the possible drivers that influence the relationships within. Changes in agri-food systems in the developing world are rapid, and are prone to become unsustainable in the long run, in the absence of policies and interventions with systemic impacts. A good example is the obesity crisis in the United States and reduction in the variety of crops (or increase in monocropping) and livestock production in the country over time.
The need of the hour is to understand, characterize and monitor these fast-paced changes, and use them to support policy and interventions in the agri-food systems. The data landscape of agriculture and the agri-food system in the developing world is data-poor, and wherever there are traces of data, they are extremely disaggregated.
Part of this problem of developing data-driven solutions to monitor these highly dynamic agri-food systems has been solved through the use of large scale mobile-based survey data collection, improvements in remote sensing and data analytics. While the large scale mobile-based surveys are bottom-up and the remote sensing-based approaches are top-down, there is still a gap between the two, that prevents high resolution spatial and temporal monitoring of these systems.
Pulse Lab Jakarta has been at the forefront of using non-conventional data sources and analytics to develop data-driven solutions for various stakeholders, but thus far the focus has largely been on sectors other than agriculture. The CGIAR Platform for Big Data in Agriculture represents that largest consortium of partners and institutions that are involved in bringing big data revolution to transform agri-food systems, largely through making agri-food system-related data FAIR (Findable, Accessible, Interoperable and Reusable), through the development of data products such as GARDIAN, and improved use of advanced remote sensing data sources and analytics — however, as a platform, have not yet explored non-conventional data sources.
This is where PLJ and the Platform now intersect. As part of this secondment from the Big Data Platform to PLJ, my role will be to work closely with the Pulse Lab team to build use cases specifically in agri-food systems using non-conventional data sources and derive insights, which can be used to monitor changes. The next few months will be exciting as both PLJ and the Platform will venture towards exploring the use of pseudonymised mobile network data, internet traffic data, social media data, through joint projects and research dives, and prove their worth within the existing data landscape of agri-food systems. We will be drawing the expertise and collaboration from the UN specialised agri-food agencies as well as government ministries and development partners. I’m excited to be at Pulse Lab Jakarta, it’s only a few weeks but it already feels like home. This is an exciting collaboration and we will share more on our joint activities as we move forward so please do stay tuned.
Pulse Lab Jakarta is grateful for the generous support from the Government of Australia.
