Farm-e represents my first foray into developing decision support tools for agriculture. Talking to growers in agricultural areas of Fiji (and also Cambodia and Timor Leste) has given me insights into the numerous challenges they face. I’ve only become more convinced of the relevance and need to develop more tools in this category, as a means of helping people lift themselves out of poverty.
It is impossible to have a conversation with growers without mention of the weather and its impact on determining their fortunes. Reliable information on weather data is difficult to access. Initially, our office went to the Fiji Meteorological Service in an attempt to get weather data only to be dismayed by the prohibitive and lengthy process involved in acquiring information that should rightly be in the public domain.
The mandate for met services like the National Weather Service in the United States is to provide consistent and reliable weather data to service providers. The result is a thriving ecosystem of companies involved in working with this data to make it available to the wider public.
Some time after this, with this problem in the back of my mind, I got in touch with a friend who I know does work in this area. Peter Shin has been a researcher at the San Diego Supercomputer Center (SDSC) at the UC San Diego campus in La Jolla for more than ten years. His recent work includes hardware and software development for remote weather sensor networks. He has been instrumental to deployments of such systems in Taiwan. I suggested to him certain agriculture-specific applications of his system that would benefit emerging market contexts—as a means to strengthen weather information infrastructure in countries where this information is not efficiently delivered.
The discussion sparked mutual interest and we promised to keep each other updated on developments.
As it turns out, restrictive policies toward access to data (in a time of growing uncertainty from climate change) are just one side of the story. On the technical side, we see that closed and proprietary systems developed by companies such as Campbell Scientific make it very costly to gather weather data—and when acquired, difficult to share.
Peter’s work involves building open source tools that builds upon software development already done by the Open Source Data Turbine Initiative. The work has been funded by the Moore Foundation and based out of the SDSC. The resulting hardware and software systems have been deployed in different environments. But, the result is a low-cost, flexible, open, and highly functional system for collecting and sharing weather-related data.
It seemed logical to me to try to bring this kind of innovation to Fiji. So, the plan is to go forward with a deployment of such a system for Fiji. We’re both excited by this prospect.
In western nations, government weather services generally provide increasingly complex weather data for free to agriculturalists and other service providers. In fact, there exists a healthy ecosystem of information providers who utilize this information to offer financial and other products to their customers. The recent $1.1 billion acquisition of Climate Corporation by Monsanto highlights this connection between improved weather data and financial products aimed at farmers.
But is Development the right pathway for pursuing this initiative? I hope to be able to tell you in a few months.
The gestation period for such projects can be quite lengthy and results are mixed. I can say this much. Peter and I are trying to compress the deployment of a very complex system that collects and reports data from sites located in numerous locations all over the country and bring in private sector support for this venture into the extremely short timespan of just a few months. This last element is absolutely crucial for the long-term viability of this project.
This site will feature continuing updates on our progress over the next few months.