Date of Award

Fall 2014

Degree Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Lori A. Hoagland

Committee Chair

Lori A. Hoagland

Committee Member 1

Ronald F. Turco

Committee Member 2

Bruce M. Applegate

Committee Member 3

Bruce P. Bordelon


Fresh market vegetable production in the Midwest U.S. is growing rapidly as a result of increased demand for locally-sourced food. High tunnels are becoming increasingly popular among Midwest vegetable growers as these structures offer many production benefits including season extension. However, vegetables require substantial fertility inputs, particularly nitrogen (N), in order to meet desired levels of productivity, and some fertility management strategies can negatively impact soil and environmental quality. This can be particularly problematic in high tunnels, where longer production periods and modification of the growing environment put these systems at greater risk for soil degradation. Identification of fertility practices that maintain soil quality and reduce negative environmental impacts is needed to improve the longevity and sustainability of these systems. The biological mechanisms controlling N cycling in Midwest vegetable production systems are not well characterized, and understanding how fertility practices alter this process are critical for limiting N loss in this region. The specific objectives of this research are to 1) quantify impacts of inorganic and organic fertility amendments on soil quality and crop productivity in a high tunnel and open field production systems, 2) determine how fertility amendments alter the structure of microbial communities that regulate N cycling processes, and 3) determine how changes in microbial community structure influence potential nitrification and denitrification activity. Urea, chicken litter, green manure (fall seeded hairy vetch (Vicia villosa Roth) plus alfalfa meal), and an unfertilized control were repeatedly applied in a high tunnel and adjacent open field plot at the Meigs Horticulture farm over a period of three growing seasons (2011 to 2013). Swiss chard (Beta vulgaris L.) was grown in 2011 and sweet peppers (Capsicum annuum L.) were grown in 2012 and 2013. Crop yield, soil pH, electrical conductivity (EC), total carbon, permanganate oxidizable carbon (POXC), plant available N, total N, and microbial activity (FDA) were quantified during each growing season, and total nutrient analysis was performed at the conclusion of the field trial. In 2013, potential nitrification (PNA) and denitrification (PDA) activity, abundance of ammonia-oxidizing archaea (AOA), bacteria (AOB), nitrite reductases (nirK and nirS) and nitrous oxide reductase (nosZ) genes, and community structure of bacterial ammonia monooxygenase (amoA) and nirK were quantified in bulk and rhizosphere soil samples collected from open field plots. In 2011, chard yield was lower in the high tunnel than the open field system, while pepper yield was greater in the high tunnel in 2012 and 2013. Chard yield was lower in plants receiving green manure than those receiving urea in 2011, but no differences in pepper yield were observed in response to fertility treatments in 2012 or 2013. Chicken and green manure treatments increased soil quality relative to urea and control treatments as indicated by greater FDA and POXC. However, EC was greater in chicken and green manure treatments than the unfertilized control, though levels were not considered inhibitory to chard or peppers. Potential nitrification and denitrification activity were greater in the rhizosphere than bulk soil, but nosZ abundance was lower in rhizosphere. Fertility treatments had a stronger impact on PNA, PDA, and N cycling gene abundances in bulk rather than rhizosphere soil. However, when averaged across bulk and rhizosphere fractions, there were no differences in PNA or AOA and AOB abundance. In contrast, green manure treatments had greater PDA in comparison to soils amended with urea and the control, though no differences in nirK, nirS, or nosZ abundance were observed between fertilized treatments. The abundance of AOB and nirK were strongly correlated with PNA and PDA, respectively, suggesting these functional groups more tightly control nitrification and denitrification processes in this dominant Midwestern soil. Multivariate analyses of AOB and nirK community profiles revealed that soil fraction and fertility treatments distinctly shaped these groups, and impacted their diversity. In conclusion, use of organic fertility amendments improve soil quality relative to inorganic amendments in intensively managed Midwest vegetable production systems, but could result in increased N loss from these systems when conditions are suitable for denitrification, though additional studies are needed to confirm denitrification efficiency. Nitrogen cycling dynamics are modified in the rhizosphere in comparison to bulk soil, which could influence N loss from these systems and therefore should be considered in soil N cycle models.