Master of Science, The Ohio State University, 2024, Plant Pathology

Soybean cyst nematode (SCN), Heterodera glycines, remains the most damaging soybean pathogen in North America. Management decisions of SCN begins with accurate soil sampling to assess current population densities. Although widely used, manual sampling is time-consuming, less effective over large areas, and may overlook spatial aggregation. This project aimed to optimize and compare the accuracy and efficiency of an automated soil sampling approach using an automated precision soil sampler (APSS) against manual soil sampling method. Three SCN-infested fields, one from Clark County (WARS) and two from Fulton (-1 and -2) County in Ohio, were selected for this study. Both manual and automated methods were used to collect soil samples following a grid pattern. The two sampling methods were evaluated using two approaches: direct SCN egg count and pixel-based counting derived from surface interpolation predicted by inverse distance weighting (IDW) model. Lin's concordance correlation coefficient (ρc) was used to assess agreement between the two methods. The egg count approach showed positive ρc values in WARS (ρc = 0.53) and Fulton-1 (ρc = 0.77), indicating agreement between manual and automated sampling methods. However, a lower correlation was observed for Fulton-2 (ρc = 0.47). Using the IDW interpolation approach, which accounts for spatial aggregation, both sampling methods detected similar SCN values (ρc = 0.99) for all fields. Moreover, removing 50% of the automated sampling points resulted in similar spatial resolution of SCN distribution across the fields, offering a significant cost-benefit advantage. We also demonstrated the feasibility of conducting on-farm trials using this automated approach. This research demonstrates that integrating automated sampling with a geospatial approach can enhance our understanding of SCN distribution, streamline soil sampling, and identify high-risk areas of infestation, thereby assisting growers in making informed decisions to ma (open full item for complete abstract)

Committee: Horacio Lopez-Nicora (Advisor); Christopher Taylor (Committee Member); Pierce Paul (Committee Member) Subjects: Plant Pathology

Master of Science, The Ohio State University, 2019, Horticulture and Crop Science

Soybean agronomic optimum seeding rate (AOSR) varies from less than 200,000 seeds ha-1 to over 400,000 seeds ha-1 based on yield potential and environmental factors, and planting at or near the AOSR helps farmers maximize yield. Understanding where AOSR is likely to be high or low is useful for soybean farmers utilizing variable rate seeding. An AOSR representing an area smaller than a whole field is referred to as local ideal seeding rate (LISR). The objective of this on-farm study was to identify soil and terrain attributes that were most predictive of differences in LISR. Randomized, replicated seeding rate strip trials were established at 4 fields in 2017 and 3 fields in 2018. Yield data taken from yield monitors were used to estimate LISR 33 to 68 times per field. Soil physical and chemical properties were measured across the field using 0.2 hectare grid samples. In order to estimate soil fertility at the same scale as LISR, geographically weighted regression and random forest interpolation methods were compared. Geographically weighted regression (GWR) had lower root mean square error and better identified low-phosphorous areas of the field, so GWR was used to interpolate all soil properties. Terrain attributes calculated from 0.76 m digital elevation models were also summarized to this scale. Random forest analysis was performed to identify which soil and terrain attributes were most important for predicting LISR within each site-year. Terrain attributes were generally more important than soil properties at all site-years. Univariate linear models were used to relate the most important soil and terrain attributes to LISR. Valley depth was an important variable for model stability in multiple sites and had a strong univariate relationship with LISR across 7 site-years. Moving from the lowest valley to the highest ridge was associated with an LISR increase of 76,000 seeds ha-1. Aspect and relative slope position also had large univariate impacts on LISR. While (open full item for complete abstract)

Committee: Laura Lindsey (Advisor); John Fulton (Committee Member); Elizabeth Hawkins (Committee Member); Pierce Paul (Committee Member); Sakthi Subburayalu (Committee Member) Subjects: Agronomy; Soil Sciences

Master of Environmental Science, Miami University, 2017, Environmental Sciences

The overall mission of the Ohio Environmental Protection Agency's Division of Surface Water is to ensure compliance with the Clean Water Act by restoring and maintaining the ecological integrity of Ohio's rivers, streams, and wetlands through biological monitoring, permitting, enforcing laws, and enhancing scientific methodology. Within the Division of Surface Water, the Wetland Ecology Group conducts wetland research and develops biocriteria and water quality standards for Ohio wetlands. This report provides a brief description of my duties as a Wetland Assessment Intern with the Wetland Ecology Group during the 2015 and 2016 field seasons. It also provides a more in depth summary of one project that involved visiting 10 natural riparian forests, collecting vegetation and soil data, and conducting a preliminary analysis of this data to help develop performance standards for stream mitigation projects. The standards may be incorporated into the Ohio EPA's Section 401 stream mitigation requirements in the future.

Committee: Suzanne Zazycki (Advisor); Sarah Dumyahn (Committee Member); Hays Cummins (Committee Member) Subjects: Environmental Science