Sigma Xi Poster Competition
Soil Health Dynamics in Kentucky Croplands and Grasslands: A Regression and Correlation Analysis
Academic Level at Time of Presentation
Graduate
Major
Masters in agriculture
List all Project Mentors & Advisor(s)
Dr. Iin Handayani
Presentation Format
Poster Presentation
Abstract/Description
This study employs regression and correlation to analyze the predictive and associative relationships between soil health indicators in Kentucky croplands and grasslands, offering statistical insights to guide sustainable management. Using Pearson’s correlation matrices and linear regression models on soil data (pH, organic matter [SOM], bulk density, porosity, water retention) from six fields (corn, soybean, tobacco, beans, Miscanthus, Mexican feather grass), we quantified how key variables interact. Regression revealed SOM as a critical predictor of soil structure, with a 1% increase in SOM reducing bulk density by 0.11 g/cm³ (β = -0.11, R² = 0.86, p < 0.01). Macroporosity exhibited a near-perfect linear relationship with soil water holding capacity (SWHC; β = 1.06, R² = 0.98, p < 0.001), where a 10% increase in macropores enhanced SWHC by 10.6%. Soil pH and SOM were strongly correlated (r = 0.89), with regression confirming a 0.36-unit pH rise per 1% SOM gain (β = 0.36, p = 0.019), aligning with grasslands’ near-neutral pH (6.61) versus acidic tobacco soils (5.41). Negative correlations dominated soil compaction dynamics: bulk density inversely correlated with total porosity (r = -1.00) and SWHC (r = -0.98), emphasizing compaction’s threat to water retention. While grasslands outperformed croplands in SOM, porosity, and pH, regression models prioritized SOM and macroporosity as actionable levers for soil health. These findings advocate for integrating perennial grasses into crop rotations and prioritizing SOM management to mitigate compaction. Limitations include short-term data and restricted crop diversity, warranting long-term, multi-region validation.
Keywords: Soil organic matter, linear regression, Pearson correlation, soil compaction, water retention, sustainable land use
Spring Scholars Week 2025
Sigma Xi Poster Competition
Soil Health Dynamics in Kentucky Croplands and Grasslands: A Regression and Correlation Analysis
This study employs regression and correlation to analyze the predictive and associative relationships between soil health indicators in Kentucky croplands and grasslands, offering statistical insights to guide sustainable management. Using Pearson’s correlation matrices and linear regression models on soil data (pH, organic matter [SOM], bulk density, porosity, water retention) from six fields (corn, soybean, tobacco, beans, Miscanthus, Mexican feather grass), we quantified how key variables interact. Regression revealed SOM as a critical predictor of soil structure, with a 1% increase in SOM reducing bulk density by 0.11 g/cm³ (β = -0.11, R² = 0.86, p < 0.01). Macroporosity exhibited a near-perfect linear relationship with soil water holding capacity (SWHC; β = 1.06, R² = 0.98, p < 0.001), where a 10% increase in macropores enhanced SWHC by 10.6%. Soil pH and SOM were strongly correlated (r = 0.89), with regression confirming a 0.36-unit pH rise per 1% SOM gain (β = 0.36, p = 0.019), aligning with grasslands’ near-neutral pH (6.61) versus acidic tobacco soils (5.41). Negative correlations dominated soil compaction dynamics: bulk density inversely correlated with total porosity (r = -1.00) and SWHC (r = -0.98), emphasizing compaction’s threat to water retention. While grasslands outperformed croplands in SOM, porosity, and pH, regression models prioritized SOM and macroporosity as actionable levers for soil health. These findings advocate for integrating perennial grasses into crop rotations and prioritizing SOM management to mitigate compaction. Limitations include short-term data and restricted crop diversity, warranting long-term, multi-region validation.
Keywords: Soil organic matter, linear regression, Pearson correlation, soil compaction, water retention, sustainable land use