Silica

Silica Report

Code

if (!requireNamespace("librarian", quietly = TRUE)) {
  # If not installed, install the package
  install.packages("librarian")
}

librarian::shelf(
  glue,
  here,
  skimr,
  ggplot2
)

data <- read.csv(here("data/df_cleaned.csv"))
parameter_name <- params$parameter_name

apply param bounds

bounds <- read.csv(here("parameter_bounds.csv"), stringsAsFactors = FALSE, strip.white = T)
lower_bound <- bounds$min[bounds$param == parameter_name]
upper_bound <- bounds$max[bounds$param == parameter_name]

filter_condition <- (data$Parameter == parameter_name & (data$Value < lower_bound | data$Value > upper_bound))

tryCatch({  # this tryCatch is for when filter_condition is logical(0) i.e. no matches
  data <- dplyr::filter(
    data, 
    !filter_condition
  )
  print(glue("{sum(filter_condition)} rows dropped as < {lower_bound} or > {upper_bound}"))
}, error = function(e){
  print(glue("no rows removed"))
})

0 rows dropped as < 0 or > 20

apply param bounds

print(glue("{sum(filter_condition)} rows dropped as < {lower_bound} or > {upper_bound}"))

0 rows dropped as < 0 or > 20

write cleaned DataFrame to a file

write.csv(data, here("data/df_cleaned_02.csv"), row.names = FALSE)

load data & skim

subset_data <- subset(data, Parameter == parameter_name)
print(skimr::skim(subset_data))

── Data Summary ────────────────────────
                           Values     
Name                       subset_data
Number of rows             22144      
Number of columns          17         
_______________________               
Column type frequency:                
  character                4          
  numeric                  13         
________________________              
Group variables            None       

── Variable type: character ────────────────────────────────────────────────────
  skim_variable n_missing complete_rate min max empty n_unique whitespace
1 Source                0             1   3  10     0        5          0
2 Site                  0             1   1  28     0      488          0
3 Parameter             0             1   6   6     0        1          0
4 Units                 0             1   4   9     0        2          0

── Variable type: numeric ──────────────────────────────────────────────────────
   skim_variable     n_missing complete_rate       mean        sd         p0
 1 ...1                      0         1     393422.    89251.    345229    
 2 Latitude                  0         1         26.0       0.805     24.2  
 3 Longitude                 0         1        -80.6       0.762    -83.6  
 4 Month                     0         1          6.68      3.43       1    
 5 Day                      28         0.999     12.5       7.58       1    
 6 Year                      0         1       2017.        6.50    1995    
 7 Value                     0         1          0.787     2.32       0    
 8 Sample.Depth             29         0.999      3.25      7.66       0    
 9 Total.Depth           19296         0.129      4.92      6.05       0.444
10 verbatimValue             0         1          0.787     2.32       0    
11 VerbatimLatitude          0         1         26.0       0.805     24.2  
12 verbatimLongitude         0         1        -80.6       0.762    -83.6  
13 Value_orig                0         1          0.787     2.32       0    
          p25        p50        p75     p100 hist 
 1 355967.    361502.    367038.    634510   ▇▁▁▁▁
 2     25.4       26.0       26.6       28.4 ▃▅▇▅▁
 3    -81.2      -80.1      -80.1      -80.0 ▁▁▁▃▇
 4      4          7         10         12   ▇▅▅▆▇
 5      6         12         18         31   ▇▇▆▃▂
 6   2017       2020       2021.      2023   ▁▁▁▁▇
 7      0.016      0.047      0.295     19.9 ▇▁▁▁▁
 8      0          0.5        2.54     247   ▇▁▁▁▁
 9      1.91       3.01       6         52.7 ▇▁▁▁▁
10      0.016      0.047      0.295     19.9 ▇▁▁▁▁
11     25.4       26.0       26.6       28.4 ▃▅▇▅▁
12    -81.2      -80.1      -80.1      -80.0 ▁▁▁▃▇
13      0.016      0.047      0.295     19.9 ▇▁▁▁▁

create params$parameter_name histogram

ggplot2::ggplot(subset_data, aes(x=Value)) +
    geom_histogram(bins=30, fill="blue", color="black") +
    scale_y_log10() +  # Transform the y-axis to a logarithmic scale
    labs(title=paste("Histogram of Values for", params$parameter_name),
         x="Value",
         y="Log Frequency") +
    theme_minimal()