Systems analysis, design, and testing for an agricultural soil compaction sensing device

Abstract

The World Resources Institute (WRI) and United Nations Food and Agriculture Organization (FAO) estimate that global agricultural food production will need to increase 56% between 2010 and 2050 to meet projected caloric demands of the growing population. Given the finite amount of global land area as well as the impacts agricultural land-use and production have on greenhouse gas emissions (GHGs), achieving this increase in consonance with climate change reduction goals presents a formidable challenge. In the past, large yield improvements have been realized from genetically modified seeds, synthetic fertilizers and pesticides, increased mechanization, and improved irrigation, but these innovations have also brought negative side effects. For example, the increasing weight of mechanized farm equipment has led to significant man-made soil compaction.

Soil compaction is the increase in bulk density, or reduction of air pore space, in a soil matrix, and it can lead to restricted root growth, poor water and nutrient infiltration, and reductions in yield. Specifically, man-made compaction has been estimated to lead to 15-20% reductions in crop yield, leading to $40-45 Billion in annual financial losses in the United States alone. Mechanical tillage is one of the most common remedies for loosening compacted soil, but the process damages soil structure and overall soil health, making it a solution that should optimally be used sparingly, only in areas where soil is severely compacted. A key challenge to enabling this, however, is compaction sensing and mapping at the field scale. In response to this challenge, a research project was undertaken through MIT Beaver Works, a collaboration with MIT Lincoln Laboratory, to explore systems-based solutions for real-time soil compaction sensing and mapping.

Through that work, a high-level system design for measuring soil compaction at the field scale was proposed based on electromagnetic sensing, including the use of ground penetrating radar (GPR) and electromagnetic induction (EMI) sensors. This thesis aims to address the highest risk aspects of the proposed approach through modelling, laboratory testing, and field testing, progressing theoretical results into increasingly more realistic settings to better understand practical limitations and potential challenges with the technical approach.