Joe Ronzio/IWMI

Using models to determine impact with small-scale irrigation

In June, the Innovation Laboratory for Small Scale Irrigation, a project under USAID’s Feeding the Future Initiative, hosted a training program in Addis Ababa, Ethiopia, for over 60 trainees from Ethiopian universities, the Ministry of Agriculture, the Ministry of Environment and Forestry, the Ministry of Water, Irrigation, and Energy, and CGIAR staff on the three main models that they will employ over a five-year project.

These models will be used to help determine the production, economic, human nutrition, and environmental impacts of introducing small-scale irrigation technologies at field, farm and watershed scales in communities in Ethiopia, Ghana and Tanzania. They include:

  • APEX, which analyzes the impacts on productivity of a particular technological interventions, such as small scale irrigation and associated practices;
  • SWAT, which analyzes the environmental impacts of these interventions at the watershed or river basin scale; and
  • FARMSIM, which analyzes the farmer’s risks of applying the technology.

The models were originally developed by the US Department of Agriculture and Texas A&M University but are being adapted through the Innovation Laboratory for Small Scale Irrigation for use in Ethiopia, Ghana and Tanzania, implemented in partnership with CGIAR centers: the International Water Management Institute (IWMI), the International Food Policy Research Institute (IFPRI) and the International Livestock Research Institute (ILRI).

For the training, IWMI used data collected on stream flow and weather variables over 3 years in Jeldu in Ethiopia’s Southwestern Abay Basin. This enabled trainees to set up models with real-world data and evaluate their accuracy.

This collaboration between CGIAR Centers and United States universities aims to conduct innovative research under field conditions and to analyze the results in a different way, using powerful models that have been developed over the years and recently validated in studies in Ethiopia over the last two years.

“We believe the combined institutional experience of the team will provide relevant results that can be applied at farm and other levels of scale,” said Neville Clarke of Texas A&M University.

“To ensure that these models are characterized to the best of our abilities, we are in the process of designing field interventions to show research into action,” said Simon Langan, Director of the IWMI East Africa Office.

Through such field interventions, researchers will be able to test if the predictions made by the models are true on the ground as well.

This will give researchers and government decision makers the confidence to use the models to predict the effects of small-scale irrigation in areas where it is not possible to conduct field experiments.

The project aligns closely with the Ministry of Agriculture’s priority to support the development of intensive agricultural systems using small scale irrigation for high value crops. With help from the models and field tests in areas like Jeldu, researchers will be able to better determine how to intensify systems, reduce erosion, and increase forage production with small scale irrigation. Specific technologies that will be tested include low cost water lifting devices; watershed management; water harvesting and drip irrigation; and irrigated fodder.

“We can conduct risk analyses for each of these technologies using data about costs and prices as well as the economic status of the farmer. We can conduct analyses that say it’s likely that adopting this new technology will improve the economic status of the farmer and reduce risk, or the opposite,” said Allan Jones of Texas A&M University.

But models aren’t the end all, be all. “Models don’t necessarily give you a definitive answer, especially in the absence of data. So we use the best data available, coupled with knowledge from local officials and farmers about the local landscape and management practices, and then combine this information in models to offer a series of alternatives and their potential impacts. But our models don’t guarantee that the impact we predict will actually happen, but instead give an indication of the likely outcomes” said Tracy Baker, Hydrologic modeler and IWMI researcher. Nevertheless, “this type of information is useful whether you’re a farmer who wants to spend money to buy a pump or if you’re a government official who wants to develop a government program to enhance the acceptance of a technology,” Jones said.

With success, the project has major implications for policy makers and investors. The project is still in its first year. Over the next four years, the models will be tested in Ethiopia, Tanzania and Ghana.