Model practice – enhancing crop models to reduce food insecurity in East Africa
A large proportion of families in East Africa struggle daily with food insecurity and malnutrition. These families are forced to subsist on as little as two hectares of land. They are restricted not only by the quantity of land, but also by the quality, since much of the soils in East Africa have low organic matter and poor soil fertility.
Farmers could increase their yields and simultaneously build richer soil in a sustainable way by adopting intensive farming practices. But, knowledge about which farming practices are the most effective is not well documented by researchers. To get a good understanding of which practice to use requires years of scientific study through trial and error and is both labor intensive and time consuming.
In more developed regions, these questions have been answered by computer models, which take far less physical effort and substantially less time to produce answers. Ideally these models could answer the same questions for farmers in East Africa. But unfortunately, most of them were developed for monoculture and conventional farming systems that do not represent the type of subsistence farming that occurs in East Africa.
To accommodate complex farming practices in East Africa, a team of CIAT-led computer modelers enhanced CropSyst, a sophisticated agricultural model, to simulate the simultaneous growth of two species and their competition for light, water, and nutrients.
While developing the model, the modelers focused on intercropping – the growth of two or more crops together at the same time – a farming practice that better reflects farming systems in East Africa.
In terms of overall yield and soil quality, intercropping offers numerous advantages and few disadvantages. Some advantages are fairly straight-forward, such as better per-area resource use (more leaf area means more light interception; more roots means more water uptake – both essential for crop growth). Other advantages are fairly subtle, such as one plant attracting beneficial insects that may help the second plant produce seeds or reduce its pest population.
Intercropping also helps sustainability by building up soil organic matter, reducing greenhouse gas emissions by providing more groundcover, and reducing nitrogen leaching and soil runoff.
On the downside, the introduction of additional crops in a finite area forces the plants to compete for light, water, and nutrients. Crop management practices can alleviate competition but some detrimental effects may remain. The yield, for example, of a crop forced to compete for resources in an intercropping system may be less than if it were grown alone.
Understanding the advantages and disadvantages of intercropping makes evident the complexity of the interactions among the crops. These types of interactions are largely lacking in many agricultural computer models making it difficult to provide farmers with best-bet crop management techniques. This is what we are attempting to address through CropSyst.
Computer models are useful because they can provide an unlimited number of “what-if” scenarios far faster and cheaper than if they were actually tested in the field. A computer model can simulate thirty years of crop growth in seconds without having to plant anything.
Through CropSyst we can provide information such as estimated yields of each crop, soil carbon and nitrogen levels, greenhouse gas emissions, and much more. By running different scenarios and comparing their results we can make better recommendations on how farmers can increase crop yields in a sustainable way.
The amount of information that comes out of the model is immense, but so is the amount of information that we need to provide for it to function properly.
To obtain the information that goes into the model, we first need to measure how crops act under certain soil and weather conditions. This is where a CIAT-led project titled “Sustainable Intensification of crop-livestock systems through improved forages” comes into play.
Part of the project involves field trials in Lushoto, Tanzania, where intercropping is widely used among farmers. Two MSc students (from Nairobi University and Sokoine University respectively) are managing Napier grass-Desmodium trials under varying levels of fertilizer.
These particular trials are interesting because they demonstrate the complex interactions and possible benefits of intercropping. Both Napier grass and Desmodium are important forage crops in the region, but Desmodium is also a nitrogen fixer. This means Desmodium converts atmospheric nitrogen into a form that makes it available in the soil. In this way, growing Desmodium as an intercrop can actually increase availability of nitrogen to the Napier grass and increase soil nitrogen in general.
The specifics of this relationship are not fully understood and the benefits may vary based on management practices. To make the interaction clear, the two MSc students are collecting a staggering amount of growth, soil, and weather data, which are being used to calibrate the intercropping model and ensure that the predictions are realistic.
Once the modelers are satisfied with the outputs of the intercropping model, they can start running “what-if” scenarios. They could, for example, adjust the time that Desmodium is planted and determine the optimum planting date relative to Napier. Or they could adjust the amount of fertilizer or tillage frequency, or other management practices in attempts to save farmers’ time and money.
In time, the model may provide farmers and policymakers with important information to help tackle the problems of food scarcity and malnutrition in East Africa.
Written by Bryan Carlson (pictured left with a Tanzanian farmer). Bryan is a Biophyiscal Modeler for Washington State University and an independent consultant.
More information about research related to intercropping with forages
The CIAT-led project is part of the Livestock and Fish CGIAR Research Program and is supported by USAID as part of the CGIAR-Unites States University Linkages Program designed to support collaborative research between US universities or USDA and CGIAR.