Kennerdell, PA Superfund Site

Load Libraries

library(dplyr)
library(stringr)
library(ggplot2)

Load data

Sediment <- read.csv("Sediment_Data_for_Test.csv")

Rename a few columns

colnames(Sediment)

##  [1] "Sample.Name"             "Location.Name"          
##  [3] "x_coord"                 "y_coord"                
##  [5] "coord_type_code"         "Matrix"                 
##  [7] "Sample.Date"             "Start.Depth"            
##  [9] "End.Depth"               "Depth.Interval"         
## [11] "Interval.Name"           "Depth.Unit"             
## [13] "Sample.Type"             "Exposure.Area"          
## [15] "Drainage.Sequence.Order" "CAS.RN"                 
## [17] "Chemical.Name"           "Result"                 
## [19] "Result.Unit"             "Result..ND.0."          
## [21] "Result..ND.QL."          "pH"                     
## [23] "Analytical.Method"       "Method.Name"            
## [25] "Method.Group"            "Qualifier"              
## [27] "Detect"                  "Detect.Number"          
## [29] "Quantitation.Limit..QL." "QL.Unit"

colnames(Sediment)[1] <- "SampleName"
colnames(Sediment)[21] <- "ResultNDQL"

Filter dataset for only the desirable components: Lead and Nickel, Depth <= 20 cm, samples that are not field duplicates

Sediment_filter <- Sediment %>% 
  filter(str_detect(SampleName, 'N')) %>% 
  filter(Start.Depth <= 0.20 & Chemical.Name %in% c("Lead", "Nickel"))

Create a boxplot of Lead and Nickel by Exposure Area

ggplot(Sediment_filter, aes(x = Chemical.Name, y = ResultNDQL)) + 
  geom_point(aes(color = factor(Detect)), size = 3, alpha = 0.5) + 
  geom_boxplot(alpha=0) + 
  facet_wrap(vars(Exposure.Area)) +
  theme_bw() +
  labs(y = "Concentration mg/kg", 
       x = "Chemical Name",
       title = "Kennerdell Superfund Site",
       color ='Detect')

Export figure

ggsave("Lead_Nickel_by_ExposureArea.jpeg", device = "jpeg", dpi=600)