Agricultural decision support framework for visualisation and prediction of Western Australian crop production

Document Type

Conference Proceeding

Publication Title

2016 3rd International Conference on Computing for Sustainable Global Development (INDIACom)

Publisher

IEEE

School

School of Science

RAS ID

23171

Comments

Armstrong, L. J., & Nallan, S. A. (2016). Agricultural decision support framework for visualisation and prediction of Western Australian crop production. In 2016 3rd International Conference on Computing for Sustainable Global Development (INDIACom) (pp. 1907-1912). IEEE. Available here

Abstract

An automated decision support system (DSS) framework is proposed to assist Western Australian farmers to improve decision making in crop variety selection for different climatic and agronomic scenarios. The AgMine DSS comprises six major components including data input, data mining, and statistical analysis, database, prediction and visualisation modules. This paper proposes how the visualization and data mining modules can be used to improve farmer decision making for individual districts. Visualisation of climate and other data was carried out using ArcGIS tool using two districts (Milling and Wongan Hills). Collation and preprocessing of data was carried out to raster files and a large repository of raster files was created for later geospatial analysis and climate and crop yield mapping. Data sets were sourced from public information resources, national and state agricultural cropping research data. Annual interpolation of rainfall and crop yield of different wheat varieties for the period 2008 to 2012 was carried out using the Arc Map and the Ordinary Kriging method with Spherical Semi-variogram models. Association rule mining techniques where also used to determine the relationship between, crop variety, rainfall and soil at district level. A framework, which combines these techniques, is proposed which can be used by farmers in decision making by providing visualisations of seasonal patterns of rainfall for individual districts and show the effect of various scenarios of dry and wet years on crop production.

Access Rights

subscription content

Link to Full Text

Share

 
COinS