Cost efficient integration of weather data into agronomic advice

In 2004 two agricultural economists developed eProd, a supplier management service, after setting up a company to export dried African Bird’s Eye chilies produced with large numbers of contracted smallholders in Kenya. At that time, there were no off-the-shelf ICT management packages available and even today it is difficult to find a complete Enterprise Resource Planner (ERP) that is affordable.

The system developed into one of the most complete and appropriate ERPs for managing agricultural supply chains with scale producers and manages contracting, field inspections, surveys, extension services, communication, mobile payment, incentives, credit, traceability, collections, etc. 

eProd comes with a field application for mobile phones or tablets and can use Bluetooth scales and printers for field collections. It can easily be integrated with hardware and software, for example, quality measurement devices for quality-based payment and specialised financial software or other ERPs. One important service of eProd is the integration of weather forecasting, production advice for farmers and field staff, and management on weather information.

Connection must be reliable and has to function in areas where internet is poor or absent. eProd is developed in Delphi and uses a MySQL database. It comes with a handheld application for Android devices. It can function in the cloud, but for most clients, internet speed is too slow. It efficiently synchronises with the network or even a simple laptop at the office. 

In 2015 the software was piloted for various sectors and eProd Solutions Ltd was registered in Kenya to commercially offer the management solution to aggregators. Even though it is now used even outside agriculture (it is used in about 20 sectors), the target users are commodity traders, farmer cooperatives and food processors in agriculture and livestock. The company has an office in Nairobi and currently, 65 users reach about 250,000 farmers in Kenya, Uganda and Tanzania. The company prepares to expand further in the region. 

Revenue model

Access to weather data is extremely important for eProd users. Advanced weather information systems exist, but the high level of expertise for the development and constant fine-tuning of crop modelling, machine learning, big data, etc. hinders the access for small-scale producers. aWhere is one of the leaders in the field of weather information for agriculture and they serve agricultural businesses across the agricultural value chain from international seed companies to innovative start-ups. However – even though they see business opportunities of serving smallholders with localised services – their traditional revenue model cannot deal with the relatively small transactions inherent of working with smallholders. 

‘From daily SMS weather forecasts, agronomic advice, like spray alerts and fertiliser advice, aggregators now have a platform to manage efficient and cost-effective the information flow between the individual producers and aWhere,’ says John Corbett, Chief Science Officer at aWhere. For example, the platform can manage commercially weather forecasts services through a subscription model, and payment can be through a check-off system after product deliveries have taken place. eProd can send SMS and email management reports such as yield forecasts in coffee for example. It can now affordably feed into aWhere models additional and highly accurate information. Via an application program interface (API) integration, eProd and aWhere exchange information which results in SMS or reports prepared for the individual farmers or managers. In addition, eProd can create an enabling environment for many more services, for example: managing weather index crop insurance, access to finance by enabling micro-financial institutions to bulk screen loan applications and manage repayment, manage contract farming, certification, household surveys, etc. All integrated in one platform.

The eProd handheld device collects the GPS locations and agronomic information such as soil type, seed variety and planting date. This information is communicated via eProd to aWhere for advanced analytics and information can now be pushed to farmers as SMS Weather forecasts, spray alerts, fertiliser advice, or email management reports with agronomic advice for field staff and management. 

Weather service for coffee farmers

A specific pilot that aWhere and eProd conduct together with coffee stakeholders in East Africa focuses on the development of weather services at coffee farmer cooperative level. Coffee is one of the most widely traded commodities in the world and millions of people depend directly or indirectly on the production and sale of coffee for their livelihoods. The global market for coffee is characterised by volatile prices and experts on the world coffee market often refer to the “coffee paradox” as lowering prices for coffee farmers result in declining incomes and profits affecting millions of people in the world’s poorest countries, while there is a coffee “boom” in consuming countries with rising sales and profits for coffee retailers and roasters. 

Coffee farmers can benefit from the management options that eProd facilitates, including certification (e.g Fair Trade, Organic, UTZ), to reach premium prices, but also through providing agronomic information services. ‘Coffee trade is all about predicting yields. The party that is best at predicting, whether it is the coffee trader, farmer cooperative or hedge fund, will make the best deals,’ says Lutz Bayerkoehler, former MD ECOM Tanzania. Models have become better at yield forecasting, but the information is typically too general for use at farmer cooperative level. An essential step will be to develop accurate and affordable localised crop monitoring and yield forecasting at the coffee farmer cooperative or local coffee trader level. If used in a smart way this can assist to build trust between producers and buyers and benefit all parties.

The coffee plant, particularly Arabica, is vulnerable to decreased production due to many factors from too warm nighttime temperatures to variable rains and drought. Taking between 30-35 weeks from flowering to harvest-ready, the warming atmosphere and increased weather variability results in the risk of stressed periods being much greater now than ever. Stressed trees are more susceptible to disease and insect damage. Stress models use numerical simulation techniques in conjunction with detailed information of soil and crop properties to evaluate the evapotranspiration deficit (Etd). The deficit, or not, and the timing (growth stage) strongly correlates to yield. 

To calculate Etd aWhere leverages potential evapotranspiration – the water demand of the environment. PET is calculated by utilising most meteorological variables and in this way aWhere understands the impact of wind, sunny skies, low humidity, and heat – exactly the conditions that impact productive coffee trees. aWhere has operationalised the FAO water balance approach to yield modelling and has also developed 20 pest and disease models.

The combination of appropriate management support systems for aggregators and access to quality weather information services impact on smallholders and can transform the landscape in which they operate.