This report is automatically generated with the R package knitr (version 1.40) .

source("R Functions/functions_QA data.R")


###there are 4 field duplicates:
  #R5AQ7102
  #R5AQ7132
  #R5AQ7185
  #R5AQ7277
    #that need to be matched with an original environmental sample

###the rest are field duplicates that have alredy been averaged. this is noted in the columns: 
  # 'Sample Type Code' (avg of field duplicates)
  # and 'Sample Comments' which lists the values of the 2 duplicates averaged

###some ND duplicates may have assumed a value of half the MDL to average


### LOAD DATA ###
R5AQ <- readxl::read_excel('Reeval_Impl_Goals_Linkage_Analysis/Data/Aqueous/R5AQ.xlsx', sheet='Aqueous Hg data Rivers R5 only', guess_max = 30000)
nrow(R5AQ) #number of rows should match the Excel file (minus the header row)
## [1] 7290
### LIST COLUMNS TO BE USED, ADD USER DEFINED COLUMNS, & RENAME COLUMNS TO CEDEN STANDARDS ###
#Use 1.READ ME.xlsx, 'ColumnsForR' to list & identify columns that match corresponding CEDEN Standard columns
keep_cols <- c('SourceID','SourceRow','Source','Project','StationName','StationCode','SampleDate','CollectionTime','LabBatch','LabSampleID','MatrixName',
               'WBT','MethodName','Analyte','Unit','Result','MDL','RL','ResultQualCode','SampleID','SampleComments','TargetLatitude',
               'TargetLongitude','QACode','BatchVerification','ComplianceCode','CollectionComments','ResultsComments','BatchComments','SampleTypeCode'
               )

temp_cols <- c('DataType') #Include columns that do not match CEDEN standards but may be useful (e.g., Unit columns for MDL & RL)
#temp_cols are removed before the data is merged with other datasets

R5AQ_new <- R5AQ %>%
  select( c(keep_cols,temp_cols) ) %>% #DO NOT CHANGE - selects columns specified above
  mutate(              #Add user defined columns
    CoordSystem = NA
    ) %>%
  rename(
    'SampleTime' = 'CollectionTime',
    'CitationCode' = 'Source'
  ) %>%
  filter(
    DataType == 'Reported' | DataType == 'Field dup',  #only keep rows where DataType is 'Reported' or 'Field dup' - these values are not part of the lab QA/QC 
    WBT != 'Not Applicable', #appears to be QC data - (only 2 occurrences when DataType == 'Reported')
    WBT != 'Well' #Well data does not concern lowland rivers or delta - visually checked that data was not applicable
  )
nrow(R5AQ_new)
## [1] 5689
#str(R5AQ_new)
#View(R5AQ_new)


### FORMAT COLUMN PARAMETERS ###
  # Standardize MatrixName Groups - "Water", "Sediment", "Soil" #
unique(R5AQ_new$MatrixName) #Identifies OLDNAMES
## [1] "samplewater"
R5AQ_new <- R5AQ_new %>%
  mutate( MatrixName = recode(MatrixName, "samplewater" = "Aqueous") )
unique(R5AQ_new$MatrixName) #New naming structure should now be listed
## [1] "Aqueous"
  # Standardize WBT (WaterBodyType) Groups - "River/Stream", "Drain/Canal", "Wetland", "Spring", "Slough", 
  #                                          "Pond",  "Lake/Reservoir", "Delta", "Forebay/Afterbay", "Not Recorded" #
unique(R5AQ_new$WBT) #Identifies OLDNAMES
##  [1] "Creek"            "Drain/Canal"      "Slough"           "River"           
##  [5] "Spring"           "Pond"             "Wetland"          "Marsh"           
##  [9] "Not Recorded"     "Lake/Reservoir"   "Canal"            "Gulch"           
## [13] "Delta"            "Forebay/Afterbay" "Tributary"
R5AQ_new <- R5AQ_new %>%
  mutate(WBT = recode(WBT,
                      "Creek" = "River/Stream",
                      "River" = "River/Stream",
                      "Marsh" = "Wetland",
                      "Canal" = "Drain/Canal",
                      "Gulch" = "River/Stream", #only applies to Harley Gulch - will be filtered out in GIS since not within lowland river or delta scope
                      "Tributary" = "River/Stream"),
         #EXAMPLE FOR WHEN "OLDNAME" is 'NA' but we want a NEWNAME - if this example is deleted, also delete the comma after "Not Recorded" above
         WBT = case_when(is.na(WBT) ~ "Not Recorded", #Use "Not Recorded" when WBT value is NA
                         TRUE ~ WBT)                  #Keep original WBT value in all other cases
         )
unique(paste(R5AQ_new$MatrixName, R5AQ_new$WBT, sep='; ')) #New naming structure for Matrix Name & WBT Groupings
##  [1] "Aqueous; River/Stream"     "Aqueous; Drain/Canal"      "Aqueous; Slough"          
##  [4] "Aqueous; Spring"           "Aqueous; Pond"             "Aqueous; Wetland"         
##  [7] "Aqueous; Not Recorded"     "Aqueous; Lake/Reservoir"   "Aqueous; Delta"           
## [10] "Aqueous; Forebay/Afterbay"
  # Standardize ResultQualCode Groups - "ND", "DNQ", NA#
unique(R5AQ_new$ResultQualCode) #Identifies OLDNAMES
## [1] "="   "ND"  "DNQ"
#[1] "="   "ND"  "DNQ" - no changes necessary


  # Format Result Column to Numeric#
  # Check column for text - based on text user needs to decide what to do
if(any(grepl('<|[a-df-zA-DF-Z]',R5AQ_new$Result))){
  old <-R5AQ_new$Result
  new <-R5AQ_new$Result
  new[grepl('<|[a-df-zA-DF-Z]', new)] <- NA #skip 'e' for exponential notation e.g., "8e-005"
  #Print what text was found and what is being done
  cat(paste0("'Result' column should be numeric but some cells contain ", grammaticList(setdiff(old, new)),
             ".\nACTIONS TAKEN:\n",
             "Numeric MDL value from SampleComments column was placed in MDL column when MDL column = '-88'.\n",
             "Result value of '<MDL' replaced with 'NA'.\n")
      )
  mdl_pattern <- ".*MDL=[[:space:]](.*)[[:space:]]ng.*"   #regex simplified: capture any character group between "MDL= " and " ng";
  #Regex explained: ".*"=any characters before; "MDL=[[:space:]]"=the string 'MDL= '; "(.*)"=capturing any character group ; "[[:space:]]ng"=characters ' ng'; ".*"=any characters after
  R5AQ_new <- R5AQ_new %>%
    mutate(
      MDL = as.character(MDL), #temporarily convert MDL column to character to insert the numeric text string from SampleComments column
      MDL = case_when(Result == "<MDL" & MDL == "-88" & grepl("ng/L", SampleComments) ~ sub(mdl_pattern, '\\1', SampleComments),  #"\\1" returns 1st pattern group; returns original string if no match
                      TRUE ~ MDL),
      MDL = as.numeric(MDL), # revert MDL column to numeric
      Result = case_when(Result == "<MDL" ~ NA_character_,
                         TRUE ~ Result),
      Result = as.numeric(new)
      )
} else {
  cat("'Result' column converted to numeric format\n")
  R5AQ_new$Result <- as.numeric(R5AQ_new$Result)
  }
## 'Result' column should be numeric but some cells contain <MDL.
## ACTIONS TAKEN:
## Numeric MDL value from SampleComments column was placed in MDL column when MDL column = '-88'.
## Result value of '<MDL' replaced with 'NA'.
#View(R5AQ_new)


  # Check if Result, MDL, & RL Columns all equal <NA> or 0 - these rows have no useful information
nrow(R5AQ_new) #Number rows before
## [1] 5689
#CODE BELOW REQUIRES USER TROUBLESHOOTING DEPENDING ON AVAILABLE COLUMNS AND SPREADSHEET SPECIFIC CONDITIONS#
R5AQ_new <- R5AQ_new %>%
  #Set 0 & negative values as blank
  mutate(Result = ifelse(Result <= 0, NA_real_, Result),
         MDL = ifelse(MDL <= 0, NA_real_, MDL),
         RL = ifelse(RL <= 0, NA_real_, RL))
na_results <- R5AQ_new %>% #Record rows where Result, MDL, & RL all equal <NA>
  filter( is.na(Result) & is.na(MDL) & is.na(RL) )
nrow(na_results)
## [1] 27
R5AQ_new <- anti_join(R5AQ_new, na_results, by='SourceRow') #returns rows from R5AQ_new not matching values in no_result
nrow(R5AQ_new) #Number rows after
## [1] 5662
  # Format Units Column - "ng/L", "mg/Kg"
unique(R5AQ_new$Unit) #Identifies OLDNAMES
## [1] "ng/L"
R5AQ_new <- R5AQ_new %>%
  standardizeUnits
unique(R5AQ_new$Unit)
## [1] "ng/L"
  # Format Date and Time Column #
# NEED TO TALK ABOUT HOW WE WANT TO DO THIS - To graph in R we need Date and Time in same column
# THE EXAMPLE CODE BELOW ASSUMES DATE AND TIME ARE IN SEPERATE COLUMNS
R5AQ_new <- R5AQ_new %>%
  #rowise() %>%    # rowise is very slow - so used sapply to make this a rowise operation
  mutate(
    #If SampleDate & CollectioTIme are not in Character format by defualt, turn it into a character class so it exports better
    SampleDate = ifelse(sapply(SampleDate, is.character), SampleDate, as.character(as.Date(SampleDate))),
    SampleTime = ifelse(sapply(SampleTime, is.character), SampleTime, format(lubridate::ymd_hms(SampleTime), "%H:%M:%S")),
    #COMBINE DATE AND TIME INTO SampleDateTime COLUMN
    SampleDateTime = ifelse(!is.na(SampleTime), paste(SampleDate, SampleTime), paste(SampleDate, '00:00:00')),
    #FORMAT SampleDateTime COLUMN TO DATE FORMAT
    SampleDateTime = lubridate::ymd_hms(SampleDateTime)
  )


### REMOVE TEMPORARY COLUMNS ###
R5AQ_new <- R5AQ_new %>%
  select(-one_of(temp_cols)) #Remove temp columns since they are no longer needed
#View(R5AQ_new)

## SAVE FORMATTED DATA AS EXCEL FILE ##
writexl::write_xlsx(R5AQ_new, path='Reeval_Impl_Goals_Linkage_Analysis/Data/Aqueous/R5AQ_ceden_format.xlsx')
# In excel, to convert SampleDate column to Date format
# 1 - Select the date column.
# 2 - Go to the Data-tab and choose "Text to Columns".
# 3 - On the first screen, leave radio button on "delimited" and click Next.
# 4 - Unselect any delimiter boxes (everything blank) and click Next.
# 5 - Under column data format choose Date, select YMD
# 6 - Click Finish.

The R session information (including the OS info, R version and all packages used):

    sessionInfo()
## R version 4.2.2 (2022-10-31 ucrt)
## Platform: x86_64-w64-mingw32/x64 (64-bit)
## Running under: Windows 10 x64 (build 22621)
## 
## Matrix products: default
## 
## locale:
## [1] LC_COLLATE=English_United States.utf8  LC_CTYPE=C                            
## [3] LC_MONETARY=English_United States.utf8 LC_NUMERIC=C                          
## [5] LC_TIME=English_United States.utf8    
## system code page: 65001
## 
## attached base packages:
## [1] grid      stats     graphics  grDevices utils     datasets  methods   base     
## 
## other attached packages:
##  [1] mgcv_1.8-41        nlme_3.1-160       lubridate_1.8.0    plotly_4.10.0     
##  [5] readxl_1.4.1       actuar_3.3-0       NADA_1.6-1.1       forcats_0.5.2     
##  [9] stringr_1.4.1      dplyr_1.0.9        purrr_0.3.4        readr_2.1.2       
## [13] tidyr_1.2.0        tibble_3.1.8       ggplot2_3.3.6      tidyverse_1.3.2   
## [17] fitdistrplus_1.1-8 survival_3.4-0     MASS_7.3-58.1     
## 
## loaded via a namespace (and not attached):
##  [1] httr_1.4.4          jsonlite_1.8.0      viridisLite_0.4.1   splines_4.2.2      
##  [5] modelr_0.1.9        assertthat_0.2.1    highr_0.9           googlesheets4_1.0.1
##  [9] cellranger_1.1.0    yaml_2.3.5          pillar_1.8.1        backports_1.4.1    
## [13] lattice_0.20-45     glue_1.6.2          digest_0.6.29       rvest_1.0.3        
## [17] colorspace_2.0-3    htmltools_0.5.3     Matrix_1.5-1        pkgconfig_2.0.3    
## [21] broom_1.0.1         haven_2.5.1         webshot_0.5.3       scales_1.2.1       
## [25] tzdb_0.3.0          googledrive_2.0.0   generics_0.1.3      ellipsis_0.3.2     
## [29] withr_2.5.0         lazyeval_0.2.2      cli_3.3.0           magrittr_2.0.3     
## [33] crayon_1.5.1        evaluate_0.16       fs_1.5.2            fansi_1.0.3        
## [37] xml2_1.3.3          tools_4.2.2         data.table_1.14.2   hms_1.1.2          
## [41] expint_0.1-7        gargle_1.2.0        lifecycle_1.0.1     munsell_0.5.0      
## [45] reprex_2.0.2        writexl_1.4.0       compiler_4.2.2      rlang_1.0.5        
## [49] rstudioapi_0.14     htmlwidgets_1.5.4   crosstalk_1.2.0     rmarkdown_2.16     
## [53] gtable_0.3.1        DBI_1.1.3           R6_2.5.1            knitr_1.40         
## [57] fastmap_1.1.0       utf8_1.2.2          stringi_1.7.8       vctrs_0.4.1        
## [61] dbplyr_2.2.1        tidyselect_1.1.2    xfun_0.32
    Sys.time()
## [1] "2023-12-29 15:59:54 PST"