Decision support systems for family farmers

How would you describe the role of ‘decision support systems’ in the context of their use for family farming?

Decision support systems include all tools and techniques that help farmers decide on their course of action for farming. Family farming faces many risks and has limited opportunities. Farming is becoming more complex and is affected by a number of factors, some within farmers’ control, but many beyond.

Indeed, this type of farming faces many constraints: participating effectively in markets, gaining access to vital inputs, such as land, water and finance, and the tools and knowledge to cope with climate change. It is important that family farmers have access to these tools, and advisory services, which they can use to forecast, plan, monitor and measure the outcomes of their decisions and actions. As far as the use of ICT-based decision support systems is concerned, these now take on many forms. Some offer forecasts, optimised plans and continuous monitoring, while others are map-based, indicating the availability and flow of resources.

My own organisation PROGIS has developed DokuPlant®, an expert database based on our own technologies and that of local experts. It contains information about farm machinery, inorganic and organic fertilizer, pesticides and chemical substances, seeds and varieties, as well as cropping-methods. These technologies optimise farmers’ returns and help reduce the risks to their farming. Expert data can be used together with other farm planning and management tools.

How can decision support systems improve family farming?

Ultimately they should help family farmers to increase their returns, reduce costs, improve the quality of their products, get better access to markets, and maintain and further improve the resilience and sustainability of the farming system and the ecology it is situated in.

What are the decision support systems available for plant protection and irrigation management for smallholder family farmers?

There are many now available. Some are commercial. Some were developed as experiments and have been offered to the public.

For plant protection we have systems that forecast, optimise interventions and help diagnose the problem. Forecasting systems obtain data from a variety of sources including automated weather systems, past data and epidemiological models. Plant protection is very complex as we cannot measure with one sensor all the conditions in which pests thrive. We have to use multiple sensors that are grouped and networked with other groups elsewhere. By using these sensors and a network of weather stations and models that predict specific situations (for example, a pest will start in five days), a farmer can start spraying in advance in an attempt to control the disease. This information can be shared with farmers in the afflicted areas by means of an SMS reminding them to ‘please spray tomorrow’, for example.

For monitoring and diagnosing pests and diseases, we use images of fields and affected crops and potential pests. In the future, it may be possible to use photometry with multiple filters, such as those used for infrared and ultraviolet wavelengths of light, and support this with pattern recognition. We will also be developing decision support systems and knowledge-based systems in conjunction with these technologies. We also need to develop the human expertise that can conduct investigations and provide recommendations with all these tools.

For irrigation, we now have map and sensor-based systems at various scales, from watersheds and farms to fields and plots. Soil moisture sensors located at different depths measure the moisture. The irrigation process can be started and managed in a precise way with groups of sensors so that a field is not flooded and water is used in an efficient way – where and when it is needed. The latest generation of sprinklers has nozzles that can be triggered individually. We are now able to use three-dimensional, accurate maps that also render elevation. With sensors linked through sensor networks, we can monitor soil humidity and local weather conditions and manage irrigation.

Precision agriculture is gradually becoming possible for smallholder family farmers. This availability is the result of more affordable technologies, including maps, sensors, the ‘Internet of Things’ and cloud computing, all of which are used to monitor and control irrigation equipment and entire systems. Video technology, drones and other intelligent farm management tools are enabling farmers to do quick and accurate modelling or simulation at very small scales, for example to calculate within a few seconds various potential outcomes for different situations and compare these outcomes for optimisation.

These services for smallholder family farmers will increase their returns and boost the resilience and sustainability of their farms in an affordable way, because the costs will be shared with others in the local community. This will bring new forms of entrepreneurship and cooperation. For example, it may lead to new information and knowledge-sharing cooperatives.

Have decision support systems that use ICTs improved the agricultural activities of family farmers?

We have reports from many farming systems all over the world. In Germany, for example, family farmers and cooperatives are benefitting from their ability to plan better. Think, for example, of farming inputs, crop monitoring and logistics, and the harvesting and processing of crops such as sugar beets. In Kenya, forecasting systems with automated weather stations have been used for pest and disease control.

What constraints are there in using decision support systems to assist family farmers?

The scale of farming influences the accessibility, affordability and ability to effectively use these systems, especially for smallholder farmers. Public extension systems in most developing countries are not set up to use these systems. Many of the systems available on public platforms are elementary and provide little ground support. There is also the question of expertise. For example, at which soil moisture levels should one switch irrigation on or off for locally grown crops on different types of soil? We need local experts who can support this technology and develop local models that will help farmers.

Another major issue is collaboration and cooperation. Take weather stations. Different organisations need similar data from weather stations for agriculture, water, tourism, rivers and risk management. But they rarely cooperate when it comes to setting up stations and sharing data. This only adds to the overall costs instead of reducing them.

We do not have a network of automatic weather stations yet, nor a service that offers data to those who want it for a small fee. In developing countries, these stations could be implemented by a governmental organisation as part of the infrastructure. The different users – public or private – can pay for the data service and build value added services for different sectors.

There is another problem as well. There is a powerful sales process at work at the moment that tells farmers what the system is capable of doing but does not reveal what the real costs are or what kind of an infrastructure is needed to support the system. So if we want to use new ICTs to support smallholder family farming, we also need new organisational models for the new services. Technology alone will not improve smallholder family farming.

How can these constraints be overcome?

Again, we have to innovate and develop new organisational models for these new services that match the rapid development of technology and meet these farmers’ needs. We also need to build the capacities of these farmers and their communities to use these technologies.

I strongly believe in more capable advisory systems. To take Austria as an example: one well-informed, experienced and practical advisor, who also understands farmers’ needs, can support 200 family farmers. We can lower costs, increase benefits and promote more ecological sustainability even in smallholder systems (and we do have them here in Austria too). The advisor is ideally part of the farming community and is paid by it. All farmers benefit! This really is achievable.

What is the future of using decision support systems to improve family farming?

Technologically, many ICT-related developments – from sensors, open data, cloud computing, the ‘Internet of Things’ and the use of drones – will contribute to more affordable, accurate, precise and available decision support systems. Decision support systems in the future need to become even more heuristic. The systems can be improved if we analyse big data and provide greater precision. They also need to improve the way they present the logic of their decision support through better visualisation for greater, more general comprehension.