Developing Agricultural Production Decision Support Simulation Tools for Increasing Agricultural Production and Food Security in Africa

Project identifier 22-NO-ACRONYM
Project Status finished
Start date 2012-01-01
End date 2015-01-01
Roadmap Theme
  • Sustainable intensification: Organizational innovations
  • Sustainable intensification: Soil, water, land and input management
  • Type of Project Applied research project
    Keyword simulation tools; decision support; production yields
    Location South Africa; Mauritius; Botswana; Kenya
    Budget 929 693
    Main Funder African Union Commission (AURG 1)
    Coordinator Durban University of Technology (DUT)
    Partners
  • University of Mauritius
  • Botswana College of Agriculture
  • Kenyatta University
  • Project Web Site https://www.dut.ac.za/iss/fsia/
    Documents AURG projects phase 1 booklet

    Description

    Background

    The Institute of Systems Science (iSS) through a multi-disciplinary research group have developed methodologies for understanding a variety of complex questions from the sciences, engineering and management. In particular we research sustainable environmental and social systems and use mathematical and computer models to predict their dynamics.

    The overarching aim of this project is to provide African farmers with decision support tools based on scientifically tested computer and mathematical models to improve agricultural production yields.

    Approach of the Project

    • Investigations into appropriate computer and mathematical models to enable yield predictions based on field and human data inputs.

    • Investigations into the pressing needs and existing practices of African farmers.

    • Development of user friendly interactive application program interface to assist farmer decision processes.

    A release plan will be used to construct an overall blueprint for the project and then the assimilation of the elements will begin. The approach will involve reviewing existing agricultural modeling platforms. Once chosen this will be dissected at source level and reconstructed on a web platform. From there a user friendly mapping structure will be set up. Both soil and rainfall databases will need to be reconstructed and integrated into the solution as they are essential. This web solution will be created with Microsoft’s MVC framework, meaning the incorporation of CSHTML, JavaScript, C#.

    The front-end of the web solution will be built from the ground up with custom animations and personalized style sheets, with Google’s mapping platform being used due to the extensive range and customizability of the API. An algorithm will be formulated to create the grid mapping of the different regions and tie in the interactions with users, using JavaScript to provide feedback from the assimilated information from the simulations. Investigations to test the usability and effectiveness of these decision support tools will be carried out.

    Major Results Achieved

    • The Agricultural Production Systems sIMulator (APSIM) program has been selected as the best resource to assist farmers.

    • We have investigated the effects of varying within season daily rainfall distributions on crop yields using simulations. Results show that within season distributions can affect yields in low rainfall seasons but this is also dependent on the use of fertilizer.

    • To improve our understanding of the impact of foliar diseases, we formulated a mathematical model based on data. Qualitative analyses were carried out and methods developed to reduce the spread of foliar diseases through effective control measures with minimum cost.

    • We have established that planting date, variety, and sowing density are better predicted by the farmer, or extension personnel, especially where computer resources and expertise are stretched.

    • Soil conditions are an important aspect of farming as they determine the crops suitable to grow and also offer a platform for the interaction of fertilizer, crops and water. Thus soil conditions have a bearing on yields.

    • Fertilizer has a clear impact in optimal soil conditions for all the rainfall conditions tested. Applying fertilizer improves yields. The relationship between soil, rain and fertilizer is more subtle in other conditions and this is where predictive models are helpful.

    Expected Impact

    Farmers can use decision support tools and models to assist in pre-season and in-season management decisions on cultivation practices, fertilization, irrigation, pesticide use, etc. This is expected to result in optimizing production and hence overall food security. It should also prevent the negative environmental effects of unnecessary agricultural inputs.