Linear regression model for predicting shallow groundwater nitrate concentration in the Black Creek Watershed, Indiana, 2022—2024

A linear regression model was developed to estimate continuous nitrate concentration in groundwater at a shallow groundwater monitoring site in the Black Creek Watershed, Indiana. A Microsoft Excel workbook was used to perform the regression analysis and estimate nitrate concentration at a USGS monitoring well (411229084541701). The regression independent variable was based on continuous (15-minute) groundwater specific conductance data that were collected at the site. The regression input dependent variable was based on nitrate concentrations in monthly discrete water quality samples that were collected between April 2022 and March 2024. Specific conductance data were averaged for each sampling day to establish matched pairs for the regression analysis. The analysis was constrained between specific conductance values of 450 and 850 uS/cm with one matched pair corresponding with a low specific conductance (<300 uS/cm) removed from the regression analysis. The regression output consists of daily groundwater nitrate concentrations during the specific conductance monitoring period between December 2021 and April 2024.