ICTworks

As mentioned in Sabeen Dhanani’s initial post in this series, How Can Sensor Technologies and Precision Farming Improve Agriculture?, technology is the easiest part of using sensors to support improved agricultural practices in developing countries. Today, fairly inexpensive sensors are readily available that can measure almost anything related to agriculture. Yet, as identified through USAID’s ICTforAg convening in Silicon Valley, numerous technology barriers persist, including:

1. Enabling efficient data transmission
2. Testing and refining analysis to accurately reflect local conditions
3. Reaching a production scale that supports a price point that enables widespread use in developing markets.

These challenges can be overcome, however, as exemplified by the Betel Meter, prototyped by IDEO.org and Proximity Designs, that helps farmers optimize their irrigation and enable precision-farming techniques in Burma.

Barrier #1: Data Transmission

The Challenge

There are a number of ways to transmit data from sensors. They can be read manually or sent via mobile data networks, SMS or over emerging radio technologies. Yet each method has its challenges. Manual readings require farmers to physically check sensors regularly. Communication over the mobile network can be expensive and requires significant power that can also be expensive and difficult to come by. New radio technologies being used for the Internet of Things (IoT) are highly efficient, but the industry has yet to standardize protocols, creating a major deterrent to investing in the infrastructure it would take to support widespread sensor use in developing countries.

The Betel Meter

With fairly small acreage, manual reads seemed sufficient for betel farmers. The initial prototype communicates information to farmers through LED lights indicated whether soil is Very Wet, Wet, Optimal, Dry or Very Dry. Initial prototypes downloaded their data to a CompactFlash (CF) card for the Proximity and IDEO.org team to analyze later, though an ideal situation (in the future) would be to include a SIM card so that farmers and agents can receive the data on their phones.

However, this will not be possible until two conditions are met: 1) mobile networks need to be reliable, which isn’t yet the case in the Dry Zone in Burma, and 2) the added cost and utility of connectivity needs to fit the service model and business plan for each context.

Barrier #2: Localized Analysis

The Challenge

Sensor data is only as good as its analysis. Sensor technology used on large commercial farms is often paired with complex models that vary by crop, soil type, geography, stage of growth and other variables that affect agricultural best practices.

Similarly, smallholder farmers need data to be crunched in a way that reflects the reality of their farm conditions and needs. As such, analytic models need to be tested and refined based on local conditions, requiring local knowledge of best practices and high-quality data collection, that can be compared against sensor data.

The Betel Meter

By focusing on a single crop type and pairing LED lights with a simple printed guide, IDEO.org and Proximity were able to create a simple tool that is calibrated to advise on optimal conditions for betel. This initial design work on the Betel Meter was part of a broader design project that explored three different types of sensors.

Building on the lessons learned from these prototypes, Proximity is continuing to develop sensor technologies that meet the needs of Myanmar’s farmers, including a moisture sensor with different ranges of sensitivity to different crop types based on their irrigation needs. Any product or service that Proximity designs and launches will draw on the expertise from a team of in-country agronomists.

Barrier #3: Accessible Price Point

The Challenge

While prices of sensors are rapidly decreasing, the cost dramatically increases with the additional components required to support power, data transmission and durability. The unit cost can come significantly down with large-scale production, but to get to a price that is affordable for the 1.5 billion smallholder farmers around the world, many of whom live on less than $2/day, the level of production would need to jump significantly. Technology providers would have to take a huge financial risk to produce devices at levels far beyond current market demand to meaningfully drop prices.

The Betel Meter

Based on very early estimates, the final product as a package was estimated to cost $15-30 at scale – while the sensor itself cost only $2. Proximity is now refining the product and developing a strategy to take a moisture sensor to market, including reducing production costs, promoting shared ownership between farmers, and finding other ways to make the tool more accessible while increasing market demand.

How Can We Accelerate Sensor Technology Advances?

Tech Providers:

• Continue to design transmission technologies that minimize power needs
• Promote interoperability of sensor and transmission technologies by working to harmonize communication protocols

Agricultural Organizations:

• Partner with technology providers to develop and test sensor technologies in developing contexts
• Utilize networks of local agricultural extension agents and experts to gather quality data that can be used to test and refine analysis for local application

Academics:

• Work with local organizations to build capacity to collect high-quality data to improve and localize data analysis capabilities
• Study impact of sensor applications and provide learning to the technology and agricultural communities

Funders:

• Bring tech providers, agricultural experts and academics together to share learning and promote partnerships
• Accelerate cost reduction by enabling production at scale (i.e., market guarantees, PPPs, and large-scale implementations by grantees/implementing partners)

This post is Part 2 of a 4-part series on the potential of low-cost sensor technologies to improve agriculture in developing countries.

1. How Can Sensor Technologies and Precision Farming Improve Agriculture?
2. 3 Barriers to Using Sensors to Improve ICTforAg
3. Managing the Data Transmission Constraints When Using ICTforAg Sensors
4. How Can We Create an Integrated ICTforAg Sensor Ecosystem?

By Jessica Heinzelman, Reach, who wrote this for Digital Development for Feed the Future, a collaboration between USAID’s Global Development Lab and Bureau for Food Security, focused on integrating a suite of coordinated digital tools and technologies into Feed the Future activities to accelerate agriculture-led economic growth and improved nutrition. More information on on low cost sensors and agriculture can be found in the Key Findings Report from the Low Cost Sensors for Agriculture workshop in June 2016.