**2. Materials and Methods**

Data used for this county-level study were obtained from three data sources: Quality Assessed Agrichemical Contaminant Nebraska Groundwater Database; Parameter– Elevation Regressions on Independent Slopes Model (PRISM) as weather data [14,15]; and water use data obtained from United States Geographical Survey (USGS).

Counties with high atrazine usage (>28.73 kg/mi2), as indicated by the National Water-Quality Assessment (NAWQA) Project, USGS (1992–2017), were included in this study. Based on this, 33 counties in the eastern Nebraska District (Burt, Butler, Cass, Cedar, Colfax, Cuming, Dakota, Dixon, Dodge, Douglas, Fillmore, Gage, Jefferson, Johnson, Lancaster, Lincoln, Madison, Nemaha, Otoe, Pawnee, Pierce, Platte, Polk, Richardson, Saline, Sarpy, Saunders, Seward, Stanton, Thayer, Washington, Wayne, York) were eligible for this study, Figure 1a,b. The findings from our recent study, which observed a potential correlation

between atrazine and estrogen-related cancers (ERC) in eastern Nebraska, further justified the selection of the study area [4].

**Figure 1.** Estimated agricultural use for atrazine in Nebraska, National Water-Quality Assessment (NAWQA) Project, United States Geological Survey, 1995. (**a**) Atrazine (EPest-low) (**b**) Atrazine (EPest-High).

While USGS pesticide usage data began in 1992, the timeframe for this study was between 1995 and 2014 due to data availability for atrazine-contaminated groundwater. The Quality Assessed Agrichemical Contaminant Nebraska Groundwater Database was queried for atrazine and its metabolites (D.E.A., D.I.A., and hydroxyatrazine) for 1995–2014. In addition to the concentration of atrazine and its metabolites in parts per billion (ppb), other variables such as well types and well depths (in feet) were also obtained. Methods used for measuring atrazine, D.E.A., and D.I.A. concentrations in the water supply wells were described elsewhere [16]. The water supply wells selected for this study were wells in eastern Nebraska counties with high atrazine usage. Numerous wells were measured for atrazine in each county, and each well was measured multiple times during the study period.

Since groundwater atrazine fate in saturated and unsaturated aquifers is greatly impacted by environmental factors, such as precipitation and soil temperature across a range of soil profiles and over time, there was a need to incorporate some of these factors into the analysis. However, since the network of land-based weather stations may lack the capacity to adequately capture the spatial variability of weather variables across the counties mentioned above, the PRISM weather dataset was used as an alternative in this study. The PRISM is a high-resolution weather dataset based on a spatial resolution of 4 km. Daily time series data for precipitation and mean air temperature were extracted from 1995 to 2014. Although soil temperature would be more critical to the kinetics of atrazine in groundwater than the air temperature, the lack of direct measurements of soil temperature resulted in the use of annual mean air temperature as a proxy for the soil temperature at depths where groundwater wells would be screened. This is valid because there is a strong relationship between mean air temperature and soil temperature due to the exchange processes between them [17,18]. Groundwater temperature is usually equal to the annual mean air temperature above the ground, and it generally fluctuates narrowly (based on depth) around this mean temperature year round.

USGS via the web interface of the National Water Information System provides water usage data for different surface or groundwater types. This database was queried for water usage between 1995 and 2010 because 2014 data was unavailable. While annual groundwater usage was unavailable, the report was available every five years for the designated counties. Since domestic well usage may be a better predictor of human exposure to groundwater atrazine, the total number of people using self-supplied domestic fresh groundwater and the amount of Million gallons per day (Mgal/d) of fresh domestic groundwater withdrawn in the selected 33 counties of eastern Nebraska were obtained.

#### *Statistical Analysis*

Variables included in the analysis were either categorical or continuous variables. Descriptive analyses were performed on the categorical variables (groundwater or well type). Meanwhile, the time (in years) of sampling groundwater atrazine, D.E.A., and D.I.A., which was initially a count variable, was categorized by five year intervals. Continuous variables were atrazine, D.E.A., D.I.A. concentrations (ppb), well depth (in feet), precipitation (in millimeters), and annual mean air temperature (in degree Celsius). Descriptive statistics for these variables included mean, standard deviation, minimum and maximum values. Given the longitudinal design of this study, we used scatter plots to examine the correlations between atrazine, D.I.A., D.E.A. concentrations (ppb), and time (years).

A cluster analysis was performed to examine the effects of well depth on groundwater atrazine concentration. Similarity for each cluster was based on the negative squared Euclidean distance of both standardized atrazine and well depth, and the shared value was 20% quantile of their similarities.

All analyses were performed on SASv9.4 (S.A.S. Institute Inc. 2013. Cary, NC, USA), and plots were made on Microsoft Excel 2016 and Prism GraphPad Prism v7.03 software (GraphPad, La Jolla, CA, USA).

#### **3. Results**

#### *3.1. Descriptive Statistics of Sampled Wells, Hydrometeorological Characteristics, and Groundwater Utilization in the Eastern Nebraska Counties*

This study included six well types (commercial, domestic, irrigation, public, monitoring, and livestock wells). Domestic wells were the most represented well-type, accounting for 59% of the study wells (Figure 2a). Furthermore, irrigation (180 ft) and domestic wells (120 ft) were the deepest of all well types in the study location (Figure 2b).

The average values of atrazine, D.E.A., and D.I.A. during the entire study period (1995–2014) for all the counties were 0.17, 0.015, and 0.073 ppb, respectively. However, no value was obtained for hydroxyatrazine, another atrazine metabolite, during this period. Furthermore, the overall average well depth (129.94 ft) is similar to the average depth of domestic wells, confirming the high prevalence of domestic well types among the study wells. Interestingly, the average withdrawals of domestic groundwater were 0.90 million gallons per day (Mgal/day), and these supplied an average of 7100 people in the selected counties of eastern Nebraska based on 2010 data (Table 1).

**Figure 2.** Descriptive characteristics of well types in the selected counties at the eastern district of Nebraska obtained from quality-assessed agrichemical contaminant Nebraska groundwater database. (**a**) The prevalence of well types (1995). (**b**) Average well depth of the different well types (1995).

**Table 1.** The descriptive statistics of atrazine, its metabolites, and well depth obtained from quality-assessed agrichemical contaminant Nebraska groundwater database (1995–2014), precipitation and annual mean air temperature obtained from PRISM climate data (1995–2014), and amount of water usage obtained from water use data (2010) in the selected counties of eastern Nebraska district.


Atrazine, D.E.A., and D.I.A. concentration were higher in shallow wells (Figure 3a–c).

**Figure 3.** Atrazine and its metabolites based on well depths. (**a**) Average atrazine concentration detected in different well depths; (**b**) Average D.E.A. concentration detected in different well depths; (**c**) Average D.I.A. concentration detected in different well depth.
