Matrix_Comparisons_Microscopy.Rmd

---
title: "Microplastics Intercalibration Study Data Analysis: Matrix Comparisons, Microscopy"
author: "Southern California Coastal Water Research Project, University of Toronto & State Water Resources Control Board of California"
date: "`r format(Sys.time(), '%d %B, %Y')`"
output:
  html_document: 
    toc: true
    toc_depth: 4
    number_sections: true
    toc_float: true
---

![](www/logobanner.png)

```{r Packages, echo = FALSE, results = 'hide', error = FALSE, warning = FALSE, message = FALSE}

#Load packages
library(tidyverse)#A ton of useful functions
library(scales)#Needed for that label wrap
library(calecopal)#Needed for color scales
library(gridExtra)#Needed to combine plots into one figure
library(knitr) #Needed to display table
library(cowplot) #Needed to combine plots
library(kableExtra) #Needed for table features (i.e., scroll_box)
library(forcats) #Needed for reordering axes
library(plotrix) #Needed for std.error fxn
library(Steel.Dwass.test) #steeldwass test

```

```{r echo = FALSE, results = 'hide', error = FALSE, warning = FALSE, message = FALSE}

Full_DB <- read_csv("https://mpchecker.sccwrp.org/checker/intercal-data?table=tbl_results_master", guess_max = 10000)

Full_DB$labid <- gsub("Lab","",as.character(Full_DB$labid))
Full_DB$sampleid <- gsub("Lab","",as.character(Full_DB$sampleid))
Full_DB$particleid <- gsub("Lab","",as.character(Full_DB$particleid))
Full_DB$photoid <- gsub("Lab","",as.character(Full_DB$photoid))

Results <- Full_DB %>%
  #Removed lab D - incomparable with other data - deviation from methods
  filter(!(labid == "D"&sampletype == "CW")) %>%
  #Removed pygcms data data
  filter(!(labid == "T"&sampletype == "CW")) %>%
  filter(!(labid == "F"&sampletype == "CW")) %>%
  #Removed second set from Lab N - 2nd set of samples processed
  filter(!(labid == "N"&sampletype == "CW"&(grepl("5|6|7|8", sampleid)))) %>% 
  #Removed labs that submitted clean water data late (want to match with original clean water data analysis)
  filter(!(labid == "WW"&sampletype == "CW")) %>%
  filter(!(labid == "NN"&sampletype == "CW")) %>%
  
  rename("microscopy" = stereoscope) %>% 
  rename("nilered" = fluorescencestaining)

Lab_Info <- read_csv("https://mpchecker.sccwrp.org/checker/intercal-data?table=tbl_labinformation", guess_max = 100000) #Import .csv file containing all lab information for other matrices

#Remove "Lab" from relevant columns
Lab_Info$labid <- gsub("Lab","",as.character(Lab_Info$labid))

#Select data relating to expertise for visual microscopy
Expertise <- Lab_Info %>%
  select(labid, matrix, expertiseextraction, expertisevisualmicroscopy, expertisenilered, expertiseftir, expertiseraman, expertisepy_gcms)

Results <- Results %>% 
   #Join expertise data to results data frame
   left_join(Expertise, by = c("labid" = "labid", "sampletype" = "matrix")) 

Reference <- read_csv('https://mpchecker.sccwrp.org/checker/intercal-data?table=sample_spiking_info', guess_max = 100000) #Import .csv file containing all reference spiking information for all matrices including clean water

```

```{r Results Tidying, echo = FALSE, results = 'hide', error = FALSE, warning = FALSE, message = FALSE}

Results_Clean <- Results %>%
  #Create factor for nominal size fractions
  mutate(sizefraction = gsub("um","µm",as.character(sizefraction))) %>% 
  mutate(sizefraction_f = as.factor(sizefraction)) %>%
  #Create factor for colors
  mutate(color_f = as.factor(color)) %>%
  #Create factor for morphology
  mutate(morphology_f = as.factor(morphology)) %>% 
  #Create factor for microscopy
  mutate(microscopy_f = as.factor(microscopy)) %>% 
  #Rename and create factor for sample type
  mutate(sampletype = case_when(
    sampletype == "CW" ~ "Clean Water",
    sampletype == "FT" ~ "Fish Tissue",
    sampletype == "SD" ~ "Sediment",
    sampletype == "DW" ~ "Dirty Water")) %>% 
  mutate(sampletype_f = as.factor(sampletype)) %>%
  #Create a new column to indicate which samples are blanks - all sampleids with a 4 or 8
  mutate(blank = ifelse(grepl("4|8", sampleid),"Y", "N")) %>%
  #Distinguish between sets of samples
  mutate(set = ifelse(grepl("1|2|3|4", sampleid),1,2)) %>% 
  #Exclude augmentation data
  filter(!(labid %in% c("R","D", "L") & grepl("5|6|7|8", sampleid) & sampletype %in% c("Sediment"))) %>% 
  filter(!(labid %in% c("L") & grepl("5|6|7|8|9|10|11|12", sampleid) & sampletype %in% c("Fish Tissue")))

```

```{r Microscopy Tidying, echo = FALSE, results = 'hide', error = FALSE, warning = FALSE, message = FALSE}

# Results are selected to only include particles that were identified via microscopy. Counts for unique samples and particle IDs for total, size fraction, color and morphology are calculated and a new column is created for each.

#Calculate the number of particles for each characteristic and combination 

Results_Clean_Microscopy <- Results_Clean %>%
  #Select particles analyzed by microscopy
  filter(microscopy_f == "Yes") %>%
  group_by(sampletype_f, sampleid) %>% 
  #Total particles per sample
  mutate(total = n_distinct(sampleid, particleid)) %>% 
  ungroup() %>% 
  #Total particles by size fraction
  group_by(sampletype_f, sampleid, sizefraction_f) %>% 
  mutate(total_sizefraction = n_distinct(sampleid, particleid)) %>%  
  ungroup() %>% 
  #Total particles by color
  group_by(sampletype_f, sampleid, color_f) %>% 
  mutate(total_color = n_distinct(sampleid, particleid)) %>% 
  ungroup() %>% 
  #Total particles by morphology
  group_by(sampletype_f, sampleid, morphology_f) %>% 
  mutate(total_morphology = n_distinct(sampleid, particleid)) %>% 
  ungroup() %>% 
  #Total particles by color and morphology within size fraction
  group_by(sampletype_f, sampleid, sizefraction_f, color_f, morphology_f) %>% 
  mutate(total_color_morphology_sizefraction = n_distinct(sampleid, particleid)) %>% 
  ungroup() %>% 
  #Total particles by color and morphology
  group_by(sampletype_f, sampleid, color_f, morphology_f) %>% 
  mutate(total_color_morphology = n_distinct(sampleid, particleid)) %>% 
  ungroup() %>% 
  #Select and rename relevant type of expertise to work for fields below
  rename("expertise" = expertisevisualmicroscopy) %>% 
  replace_na(list(expertise = "Not Reported"))

```

```{r Reference Data Tidying, echo = FALSE, results = 'hide', error = FALSE, warning = FALSE, message = FALSE}

#Particle counts within each sample are totaled for the following categories: total, size fraction, morphology, color, and polymer. A new column is created for each categories. Values are sample specific. 

Reference_Clean <- Reference %>%
  #Remove small size fraction spiking data from lab R that isn't in other samples
  filter(!(labid == "R" & sizefraction == "20-Jan" & sampletype == "DW")) %>% 
  filter(!(labid == "R" & sizefraction == "20-212" & sampletype == "DW")) %>% 
  
  #Drop spiking values of 0 or NAs
  filter(!averagenumber == 0) %>% 
  
  #Create factor for color
  mutate(color = case_when(
    color == "white" ~ "White",
    color == "orange" ~ "Orange",
    color == "green" ~ "Green",
    color == "red" ~ "Red",
    color == "blue" ~ "Blue",
    color == "clear" ~ "Clear")) %>% 
  mutate(color_f = as.factor(color)) %>% 
  #Create factor for shape
  mutate(morphology = case_when(
    morphology == "fragment" ~ "Fragment",
    morphology == "sphere" ~ "Sphere",
    morphology == "fiber" ~ "Fiber")) %>%  
    mutate(morphology_f = as.factor(morphology)) %>% 
  #Create factor for size fraction
   mutate(sizefraction = case_when(
    sizefraction == "20-Jan" ~ "1-20 µm",
    sizefraction == "1-20" ~ "1-20 µm",
    sizefraction == "20-212" ~ "20-212 µm",
    sizefraction == "212-500" ~ "212-500 µm",
    sizefraction == ">500" ~ ">500 µm")) %>%   
  mutate(sizefraction_f = as.factor(sizefraction)) %>%
  mutate(plastic_f = as.factor(plastic)) %>%
  mutate(sampletype = case_when(
    sampletype == "CW" ~ "Clean Water",
    sampletype == "FT" ~ "Fish Tissue",
    sampletype == "SD" ~ "Sediment",
    sampletype == "DW" ~ "Dirty Water")) %>% 
  mutate(sampletype_f = as.factor(sampletype)) %>%
  
#All of the following calculations distinguish between plastic and non-plastic spikes
  #Total particle spike per sample
  group_by(sampletype_f, sampleid, plastic_f) %>% 
  mutate(total_spike = sum(averagenumber, na.rm = TRUE)) %>% 
  ungroup() %>% 
  #Total particle spike per size fraction
  group_by(sampletype_f, sampleid, plastic_f, sizefraction_f) %>% 
  mutate(total_spike_sizefraction = sum(averagenumber, na.rm = TRUE)) %>% 
  ungroup() %>%
  #Total particle spike per color
  group_by(sampletype_f, sampleid, plastic_f, color_f) %>% 
  mutate(total_spike_color = sum(averagenumber, na.rm = TRUE)) %>%  
  ungroup() %>%
  #Total particle spike per morphology
  group_by(sampletype_f, sampleid, plastic_f, morphology_f) %>% 
  mutate(total_spike_morphology = sum(averagenumber, na.rm = TRUE)) %>% 
  ungroup()%>%
  #Total particle spike per size fraction, color and morphology
  group_by(sampletype_f, sampleid, plastic_f, sizefraction_f, color_f, morphology_f) %>% 
  mutate(total_spike_color_morphology_sizefraction = sum(averagenumber, na.rm = TRUE)) %>% 
  ungroup() %>% 
  #Total particle spike per color and morphology
  group_by(sampletype_f, sampleid, plastic_f, color_f, morphology_f) %>% 
  mutate(total_spike_color_morphology = sum(averagenumber, na.rm = TRUE)) %>% 
  ungroup() 

```

```{r echo = FALSE, error = FALSE, warning = FALSE, message = FALSE}

#Join relevant columns of reference data frame and results to compare expected counts to actual counts for microscopy only data. Recovery for each category is calculated on a sample-specific basis. A new recovery column is created for each category. 

#Total
totalspikecol <- Reference_Clean %>% #Select total_spike column from reference
  distinct(sampletype_f, sampleid, plastic_f, total_spike)
  
Join_Microscopy_Total <- inner_join(Results_Clean_Microscopy, totalspikecol, by = c("sampleid", "sampletype_f")) %>% #Join total_spike column to microscopy results data frame
    mutate(total_pct_recovery = ((total/total_spike)*100)) %>%  #Create a new column for the total recovery of microscopy only results
    group_by(labid, set, sampletype_f, sampleid, plastic_f, microscopy_f, total, total_spike, total_pct_recovery, expertise) %>% 
    summarise()
    
#Size Fraction
sizefractionspikecol <- Reference_Clean %>% 
  distinct(sampletype_f, sampleid, plastic_f, sizefraction_f, total_spike_sizefraction)
  
sizefractionsumresults <- Results_Clean_Microscopy %>% 
  filter(!grepl("4", sampleid)) %>%
  filter(!grepl("8", sampleid)) %>% #Dropping blank samples from results
  drop_na(sizefraction_f) %>% #Drop particles not assigned to a size fraction
  group_by(labid, set, sampletype_f, sampleid, microscopy_f, sizefraction_f, total_sizefraction, expertise) %>% 
  summarise()

Join_Microscopy_SizeFraction <- inner_join(sizefractionsumresults, sizefractionspikecol, by = c("sampleid" = "sampleid", "sizefraction_f" = "sizefraction_f", "sampletype_f")) %>%  
    mutate(sizefrac_pct_recovery = ((total_sizefraction/total_spike_sizefraction)*100)) 

#Color
colorspikecol <- Reference_Clean %>% 
  distinct(sampletype_f, sampleid, plastic_f, color_f, total_spike_color)
  
colorsumresults <- Results_Clean_Microscopy %>% 
  filter(!grepl("4", sampleid)) %>%
  filter(!grepl("8", sampleid)) %>% #Dropping blank samples from results
  group_by(labid, set, sampletype_f, sampleid, microscopy_f, color_f, total_color, expertise) %>% 
  summarise()

Join_Microscopy_Color <- inner_join(colorsumresults, colorspikecol, by = c("sampleid" = "sampleid", "color_f" = "color_f", "sampletype_f")) %>%  
    mutate(color_pct_recovery = ((total_color/total_spike_color)*100)) 

#Morphology
morphologyspikecol <- Reference_Clean %>% 
  distinct(sampletype_f, sampleid, plastic_f, morphology_f, total_spike_morphology)
  
morphologysumresults <- Results_Clean_Microscopy %>% 
  filter(!grepl("4", sampleid)) %>%
  filter(!grepl("8", sampleid)) %>% #Dropping blank samples from results
  group_by(labid, set, sampletype_f, sampleid, microscopy_f, morphology_f, total_morphology, expertise) %>% 
  summarise()

Join_Microscopy_Morphology <- inner_join(morphologysumresults, morphologyspikecol, by = c("sampleid" = "sampleid", "morphology_f" = "morphology_f", "sampletype_f")) %>%  
    mutate(morphology_pct_recovery = ((total_morphology/total_spike_morphology)*100))

#Combination - Color and Morphology Only

 combocmspikecol <- Reference_Clean %>%
   distinct(sampletype_f, sampleid, plastic_f, color_f, morphology_f, total_spike_color_morphology)

 combosumresultscm <- Results_Clean_Microscopy %>%
   filter(!grepl("4", sampleid)) %>%
   filter(!grepl("8", sampleid)) %>% #Dropping blank samples from results
   group_by(labid, set, sampletype_f, sampleid, microscopy_f, color_f, morphology_f, total_color_morphology, expertise) %>%
   summarise()

 Join_Microscopy_Combo_CM <- inner_join(combosumresultscm, combocmspikecol, by = c("sampleid" = "sampleid", "morphology_f" = "morphology_f", "color_f" = "color_f", "sampletype_f")) %>%
   mutate(color_morphology_pct_recovery = ((total_color_morphology/total_spike_color_morphology)*100))


#Combination - Size Fraction, Color and Morphology 

 combospikecol <- Reference_Clean %>%
   distinct(sampletype_f, sampleid, plastic_f, sizefraction_f, color_f, morphology_f, total_spike_color_morphology_sizefraction)

 combosumresults <- Results_Clean_Microscopy %>%
   filter(!grepl("4", sampleid)) %>%
   filter(!grepl("8", sampleid)) %>% #Dropping blank samples from results
   drop_na(sizefraction_f) %>% #Drop particles not assigned to a size fraction
   group_by(labid, set, sampletype_f, sampleid, microscopy_f, sizefraction_f, color_f, morphology_f, total_color_morphology_sizefraction, expertise) %>%
   summarise()

 Join_Microscopy_Combo <- inner_join(combosumresults, combospikecol, by = c("sampleid" = "sampleid", "morphology_f" = "morphology_f", "color_f" = "color_f", "sizefraction_f" = "sizefraction_f", "sampletype_f")) %>%
   mutate(color_morphology_sizefraction_pct_recovery = ((total_color_morphology_sizefraction/total_spike_color_morphology_sizefraction)*100))

```

# Spiking Summary

```{r echo = FALSE, error = FALSE, warning = FALSE, message = FALSE}

#total spike
a <- Reference_Clean %>%
  distinct(sampletype_f, sampleid, plastic_f, total_spike) %>% 
  group_by(sampletype_f, plastic_f) %>% 
  mutate(total_spike_mean = mean(total_spike)) %>% 
  mutate(total_spike_sd = sd(total_spike)) %>%
  ungroup() %>% 
  group_by(sampletype_f, plastic_f, total_spike_mean, total_spike_sd) %>% 
  summarise() %>%
  rename("Matrix" = sampletype_f, "Plastic?" = plastic_f, "Number of Spiked Particles" = total_spike_mean, "Standard Deviation" = total_spike_sd)

kable(a, digits = 2, caption = "Average Number of Spiked Particles per Sample") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px") 

#size fraction
b <- Reference_Clean %>%
  distinct(sampletype_f, sampleid, plastic_f, sizefraction_f, total_spike_sizefraction) %>% 
  group_by(sampletype_f, plastic_f, sizefraction_f) %>% 
  mutate(total_spike_mean = mean(total_spike_sizefraction)) %>% 
  mutate(total_spike_sd = sd(total_spike_sizefraction)) %>%
  ungroup() %>% 
  group_by(sampletype_f, plastic_f, sizefraction_f, total_spike_mean, total_spike_sd) %>% 
  summarise() %>%
  rename("Matrix" = sampletype_f, "Plastic?" = plastic_f, "Size Fraction" = sizefraction_f, "Number of Spiked Particles" = total_spike_mean,
         "Standard Deviation" = total_spike_sd)

kable(b, digits = 2, caption = "Average Number of Spiked Particles per Sample by Size Fraction") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px") 

#color
c <- Reference_Clean %>%
  distinct(sampletype_f, sampleid, plastic_f, color_f, total_spike_color) %>% 
  group_by(sampletype_f, plastic_f, color_f) %>% 
  mutate(total_spike_mean = mean(total_spike_color)) %>% 
  mutate(total_spike_sd = sd(total_spike_color)) %>%
  ungroup() %>% 
  group_by(sampletype_f, plastic_f, color_f, total_spike_mean, total_spike_sd) %>% 
  summarise() %>%
  rename("Matrix" = sampletype_f, "Plastic?" = plastic_f, "Color" = color_f, "Number of Spiked Particles" = total_spike_mean, 
          "Standard Deviation" = total_spike_sd)

kable(c, digits = 2, caption = "Average Number of Spiked Particles per Sample by Color") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px") 

#morphology
d <- Reference_Clean %>%
  distinct(sampletype_f, sampleid, plastic_f, morphology_f, total_spike_morphology) %>% 
  group_by(sampletype_f, plastic_f, morphology_f) %>% 
  mutate(total_spike_mean = mean(total_spike_morphology)) %>%
  mutate(total_spike_sd = sd(total_spike_morphology)) %>%
  ungroup() %>% 
  group_by(sampletype_f, plastic_f, morphology_f, total_spike_mean, total_spike_sd) %>% 
  summarise() %>%
  rename("Matrix" = sampletype_f, "Plastic?" = plastic_f, "Morphology" = morphology_f, "Number of Spiked Particles" = total_spike_mean, 
         "Standard Deviation" = total_spike_sd)

kable(d, digits = 2, caption = "Average Number of Spiked Particles per Sample by Morphology") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px") 

#morphology & color within size fraction
e <- Reference_Clean %>%
  distinct(sampletype_f, sampleid, plastic_f, morphology_f, color_f, sizefraction_f, total_spike_color_morphology_sizefraction) %>% 
  group_by(sampletype_f, plastic_f, morphology_f, color_f, sizefraction_f) %>% 
  mutate(total_spike_mean = mean(total_spike_color_morphology_sizefraction)) %>% 
  mutate(total_spike_sd = sd(total_spike_color_morphology_sizefraction)) %>%
  ungroup() %>% 
  group_by(sampletype_f, plastic_f, morphology_f, color_f, sizefraction_f, total_spike_mean, total_spike_sd) %>% 
  summarise() %>%
  rename("Matrix" = sampletype_f, "Plastic?" = plastic_f, "Morphology" = morphology_f, "Color" = color_f, "Size Fraction" = sizefraction_f, 
         "Number of Spiked Particles" = total_spike_mean, "Standard Deviation" = total_spike_sd)

kable(e, digits = 2, caption = "Average Number of Spiked Particles per Sample by Morphology & Color Within Size Fraction") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

```

# Counts

The following section describes data as absolute particle counts that were reported in each sample. Note that Dirty Water samples were not spiked with any particles less than 212µm and are therefore distinct from the other matrices.  

```{r echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, fig.width = 12, fig.height = 10}

spike <- Join_Microscopy_Total %>%
  #Remove spiked particles with problematic RSD
  filter(!(labid == "R" & sampletype_f == "Clean Water")) %>% 
  filter(plastic_f != "N") %>% 
  ungroup() %>%
  group_by(sampletype_f) %>% 
  mutate(mean_spike = mean(total_spike)) 

#Total - By Lab

a <- Results_Clean_Microscopy %>% 
  #Data manipulation
  filter(!grepl("4", sampleid)) %>%
  filter(!grepl("8", sampleid)) %>% #Dropping blanks
  group_by(labid, sampleid, total, expertise, sampletype_f) %>% 
  summarise() %>% 
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%   
  #Creating plot
  ggplot(aes(x= labid, y= total)) +
  theme_test() +
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(data = spike, aes(yintercept = mean_spike), linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  facet_wrap(~sampletype_f, ncol = 2)+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Particle Count", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Raw Particle Count", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line represents the number of spiked polymer particles.\nNot blank corrected.")

plot(a)

b <- Results_Clean_Microscopy %>% 
  filter(blank != "Y") %>% 
  select(sampletype_f, labid, sampleid, total) %>%
  rename("Matrix" = sampletype_f) %>%
  rename("Lab ID" = labid) %>%
  rename("Sample ID" = sampleid) %>%
  rename("Total Particles" = total) %>%
  distinct() 

kable(b, digits = 2, caption = "Raw Particle Count. Summary statistics by sample (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

#Total - Summary
c <- Results_Clean_Microscopy %>%  
  filter(!grepl("4", sampleid)) %>%
  filter(!grepl("8", sampleid)) %>% #Dropping blanks
  group_by(labid, set, sampleid, microscopy_f, total, sampletype_f) %>% 
  summarise() %>% 
  group_by(labid) %>% 
  mutate(total_mean = mean(total)) %>% 
  ungroup() %>% 
  group_by(labid, microscopy_f, total_mean, sampletype_f) %>% 
  summarise() %>%   
  #Creating plot
  ggplot(aes(x=microscopy_f, y= total_mean)) +
  theme_test() +
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_hline(data = spike, aes(yintercept = mean_spike), linetype = "dotted")+
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +
  scale_fill_manual(values = c("darkmagenta")) +
  facet_wrap(~sampletype_f, ncol = 2)+
  expand_limits(y = 0) +
  labs(y = "Particle Count", fill = "Lab", title = "Microscopy", subtitle = "Raw Particle Count", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents the number of spiked polymer particles.\nNot blank corrected.")
  
plot(c)

d <- Results_Clean_Microscopy %>% 
  filter(blank != "Y") %>% 
  group_by(sampletype_f, labid, set, total) %>% 
  summarise() %>% 
  mutate(`Mean`= mean(total)) %>%  
  mutate(`Median` = median(total)) %>%
  mutate(`SD` = sd(total)) %>%
  mutate(`Min` = min(total)) %>% 
  mutate(`Max` = max(total)) %>%
  ungroup() %>% 
  select(sampletype_f, labid, `Mean`, `Median`, `SD`, `Min`, `Max`) %>%
  rename("Matrix" = sampletype_f) %>%
  rename("Lab ID" = labid) %>%
  distinct() 

kable(d, digits = 2, caption = "Raw Particle Count. Summary statistics by laboratory (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

e <- Results_Clean_Microscopy %>%  
  filter(!grepl("4", sampleid)) %>%
  filter(!grepl("8", sampleid)) %>% #Dropping blanks
  group_by(sampletype_f, labid, set, total) %>% 
  summarise() %>% 
  mutate(total_mean = mean(total)) %>% 
  ungroup() %>% 
  distinct(sampletype_f, labid, total_mean) %>%
  group_by(sampletype_f) %>% 
  mutate(`Mean`= mean(total_mean)) %>%  
  mutate(`Median` = median(total_mean)) %>%
  mutate(`SD` = sd(total_mean)) %>%
  mutate(`Min` = min(total_mean)) %>% 
  mutate(`Max` = max(total_mean)) %>%
  ungroup() %>% 
  select(sampletype_f, `Mean`, `Median`, `SD`, `Min`, `Max`) %>%
  rename("Matrix" = sampletype_f) %>%
  distinct() 

kable(e, digits = 2, caption = "Raw Particle Count. Global summary statistics. Not blank corrected.")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

#Size Fraction Count
f <- Results_Clean_Microscopy %>%  
  filter(!grepl("4", sampleid)) %>%
  filter(!grepl("8", sampleid)) %>% #Dropping blanks
  filter(sampletype_f != "Dirty Water") %>% 
  group_by(sampletype_f, sizefraction_f, labid, set, total_sizefraction) %>% 
  summarise() %>% 
  mutate(total_mean = mean(total_sizefraction)) %>% 
  ungroup() %>% 
  distinct(sampletype_f, sizefraction_f, labid, total_mean) %>%
  group_by(sampletype_f, sizefraction_f) %>% 
  mutate(`Mean`= mean(total_mean)) %>%  
  mutate(`Median` = median(total_mean)) %>%
  mutate(`SD` = sd(total_mean)) %>%
  mutate(`Min` = min(total_mean)) %>% 
  mutate(`Max` = max(total_mean)) %>%
  ungroup() %>% 
  select(sampletype_f, sizefraction_f, `Mean`, `Median`, `SD`, `Min`, `Max`) %>%
  rename("Matrix" = sampletype_f) %>%
  rename("Size Fraction" = sizefraction_f) %>%
  distinct() 

kable(f, digits = 2, caption = "Raw Particle Count: Size Fraction. Global summary statistics. Not blank corrected.")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

#Color Count
g <- Results_Clean_Microscopy %>%  
  filter(!grepl("4", sampleid)) %>%
  filter(!grepl("8", sampleid)) %>% #Dropping blanks
  filter(sampletype_f != "Dirty Water") %>% 
  group_by(sampletype_f, color_f, labid, set, total_color) %>% 
  summarise() %>% 
  mutate(total_mean = mean(total_color)) %>% 
  ungroup() %>% 
  distinct(sampletype_f, color_f, labid, total_mean) %>%
  group_by(sampletype_f, color_f) %>% 
  mutate(`Mean`= mean(total_mean)) %>%  
  mutate(`Median` = median(total_mean)) %>%
  mutate(`SD` = sd(total_mean)) %>%
  mutate(`Min` = min(total_mean)) %>% 
  mutate(`Max` = max(total_mean)) %>%
  ungroup() %>% 
  select(sampletype_f, color_f, `Mean`, `Median`, `SD`, `Min`, `Max`) %>%
  rename("Matrix" = sampletype_f) %>%
  rename("Color" = color_f) %>%
  distinct() 

kable(g, digits = 2, caption = "Raw Particle Count: Color. Global summary statistics. Not blank corrected.")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

#Morphology Count
h <- Results_Clean_Microscopy %>%  
  filter(!grepl("4", sampleid)) %>%
  filter(!grepl("8", sampleid)) %>% #Dropping blanks
  filter(sampletype_f != "Dirty Water") %>% 
  group_by(sampletype_f, morphology_f, labid, set, total_morphology) %>% 
  summarise() %>% 
  mutate(total_mean = mean(total_morphology)) %>% 
  ungroup() %>% 
  distinct(sampletype_f, morphology_f, labid, total_mean) %>%
  group_by(sampletype_f, morphology_f) %>% 
  mutate(`Mean`= mean(total_mean)) %>%  
  mutate(`Median` = median(total_mean)) %>%
  mutate(`SD` = sd(total_mean)) %>%
  mutate(`Min` = min(total_mean)) %>% 
  mutate(`Max` = max(total_mean)) %>%
  ungroup() %>% 
  select(sampletype_f, morphology_f, `Mean`, `Median`, `SD`, `Min`, `Max`) %>%
  rename("Matrix" = sampletype_f) %>%
  rename("Morphology" = morphology_f) %>%
  distinct() 

kable(h, digits = 2, caption = "Raw Particle Count: Morphology. Global summary statistics. Not blank corrected.")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

```

# Blank Contamination

```{r echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, fig.width = 12, fig.height = 10}

#Total - By Lab
a <- Results_Clean_Microscopy %>%
  filter(!(labid == "Z"&sampletype_f == "Clean Water")) %>%
  filter(!(labid == "M"&sampletype_f == "Dirty Water")) %>%
  filter(!(labid == "M"&sampletype_f == "Fish Tissue")) %>%
  filter(!(labid == "M"&sampletype_f == "Sediment")) %>%
  #Data manipulation
  filter(blank == "Y") %>% 
  group_by(labid, sampleid, total, expertise, sampletype_f) %>% 
  summarise() %>% 
  ungroup() %>% 
  #Creating plot
  ggplot(aes(x= labid, y= total)) +
  theme_test() +
  geom_point(aes(color = sampletype_f, fill = sampletype_f),alpha = 0.7, size = 3) +
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4"))+
  scale_color_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4"))+
  facet_wrap(~sampletype_f, ncol = 2)+
  expand_limits(y = 0) +
  labs(x = "Lab", y = "Particle Count", title = "Microscopy", subtitle = "Blank Particle Count")

plot(a)

b <- Results_Clean_Microscopy %>%
  filter(!(labid == "Z"&sampletype_f == "Clean Water")) %>%
  filter(!(labid == "M"&sampletype_f == "Dirty Water")) %>%
  filter(!(labid == "M"&sampletype_f == "Fish Tissue")) %>%
  filter(!(labid == "M"&sampletype_f == "Sediment")) %>%
  filter(blank == "Y") %>% 
  select(sampletype_f, labid, sampleid, total) %>%
  rename("Matrix" = sampletype_f) %>%
  rename("Lab ID" = labid) %>%
  rename("Sample ID" = sampleid) %>%
  rename("Total Particles" = total) %>%
  distinct() 

kable(b, digits = 2, caption = "Blank Particle Count.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

blank_stats <- b %>% 
  ungroup() %>% 
  group_by(Matrix) 

# shapiro.test(log10(blank_stats$`Total Particles`)) #Normal distribution after log transformation
# bartlett.test(log10(`Total Particles`) ~ Matrix, data = blank_stats) #Equal variance after log transformation
# x <- aov(log10(`Total Particles`) ~ Matrix, data = blank_stats) #p = 0.00733
# summary(x)
# TukeyHSD(x) #Significant differences between clean and dirty water

kruskal.test(`Total Particles` ~ Matrix, data = blank_stats)
Steel.Dwass(blank_stats$`Total Particles`, as.character(blank_stats$Matrix))

#Total - Summary
c <- Results_Clean_Microscopy %>%
  filter(!(labid == "Z"&sampletype_f == "Clean Water")) %>%
  filter(!(labid == "M"&sampletype_f == "Dirty Water")) %>%
  filter(!(labid == "M"&sampletype_f == "Fish Tissue")) %>%
  filter(!(labid == "M"&sampletype_f == "Sediment")) %>%
  filter(blank == "Y") %>% 
  group_by(labid, sampleid, total, sampletype_f) %>% 
  summarise() %>% 
  ungroup() %>% 
  #Creating plot
  ggplot(aes(x=sampletype_f, y= total)) +
  theme_test() +
  geom_point(aes(color = sampletype_f, fill = sampletype_f, alpha = 0.7), size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = sampletype_f, fill = sampletype_f), outlier.shape = NA) +
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        legend.position = "none",
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4")) +
  scale_fill_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4")) +
  expand_limits(y = 0) +
  labs(y = "Particle Count", title = "Microscopy", subtitle = "Blank Particle Count")
  
plot(c)

d <- Results_Clean_Microscopy %>%
  filter(!(labid == "Z"&sampletype_f == "Clean Water")) %>%
  filter(!(labid == "M"&sampletype_f == "Dirty Water")) %>%
  filter(!(labid == "M"&sampletype_f == "Fish Tissue")) %>%
  filter(!(labid == "M"&sampletype_f == "Sediment")) %>%
  filter(blank == "Y") %>% 
  group_by(labid, sampletype_f, total) %>% 
  summarise() %>% 
  ungroup() %>%
  group_by(sampletype_f) %>% 
  mutate(`Mean`= mean(total)) %>%  
  mutate(`Median` = median(total)) %>%
  mutate(`SD` = sd(total)) %>%
  mutate(`Min` = min(total)) %>% 
  mutate(`Max` = max(total)) %>%
  ungroup() %>% 
  select(sampletype_f, `Mean`, `Median`, `SD`, `Min`, `Max`) %>%
  rename("Matrix" = sampletype_f) %>%
  distinct() 

kable(d, digits = 2, caption = "Blank Particle Count.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

```

```{r Manuscript Figure S2, echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, include = FALSE}

FigS2 <- Results_Clean_Microscopy %>%
  filter(!(labid == "Z"&sampletype_f == "Clean Water")) %>%
  filter(!(labid == "M"&sampletype_f == "Dirty Water")) %>%
  filter(!(labid == "M"&sampletype_f == "Fish Tissue")) %>%
  filter(!(labid == "M"&sampletype_f == "Sediment")) %>%
  filter(blank == "Y") %>% 
  group_by(labid, sampleid, total, sampletype_f) %>% 
  summarise() %>% 
  ungroup() %>%
  mutate(sampletype_f = case_when(
    sampletype_f == "Clean Water" ~ "Drinking Water",
    sampletype_f == "Dirty Water" ~ "Surface Water",
    sampletype_f == "Fish Tissue" ~ "Fish Tissue",
    sampletype_f == "Sediment" ~ "Sediment")) %>%
  #Creating plot
  ggplot(aes(x=sampletype_f, y= total)) +
  theme_test() +
  geom_point(aes(color = sampletype_f, fill = sampletype_f, alpha = 0.7), size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = sampletype_f, fill = sampletype_f), outlier.shape = NA) +
  theme(text = element_text(size = 20),
        plot.title = element_text(hjust = 0.5),
        legend.position = "none",
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("royalblue", "darksalmon", "goldenrod4", "forestgreen")) +
  scale_fill_manual(values = c("royalblue", "darksalmon", "goldenrod4", "forestgreen")) +
  expand_limits(y = 0) +
  labs(y = "Particle Count")
  
plot(FigS2)

saveRDS(FigS2, file = "./Manuscript_Figures/FigS2")

```


# Clean Water, Fish Tissue and Sediment Comparisons

Clean Water, Fish Tissue and Sediment samples were spiked with the same materials. Therefore, comparisons may be made for all particle types. Matrix comparisons including dirty water are included in later sections, but only for particle types >212 µm. 

## Recovery

The following plots display the recovery (reported count/spiked count (%)) obtained from microscopy. **Reported particle characteristics must match expected particle characteristics of the spiking materials to count towards recovery estimations.** For example, the colors of polymer spiking materials are defined as white, orange, blue, clear and green. Colors outside of these definitions are not included in this section (see Counts). 

The dotted line is at 100% and the grey area represents the mean RSD for each given category. The data are **not** blank corrected.

Recovery in this section is estimated based on a single particle characteristic. For example, if the sample was spiked with 10 green particles and 8 green particles were recovered, recovery is estimated as 80% regardless of whether or not the shape or size fraction also matches the spiking material. More restrictive recovery analysis are reported later on in this document.

### Total

```{r echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, fig.width = 12, fig.height = 10, fig.height = 10}

#Total - By Lab
a <- Join_Microscopy_Total %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>% 
  filter(plastic_f != "N") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%  
  #Creating plot
  ggplot(aes(x=labid, y= total_pct_recovery)) +
  theme_test() +
  geom_rect(ymin = 95.7, ymax = 104.3, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Total Particle Recovery", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(a)

#Total - By Lab
a_a <- Join_Microscopy_Total %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>% 
  filter(plastic_f != "N") %>% 
  ungroup() %>% 
  group_by(labid) %>% 
  mutate(All_Matrices = if_else(n_distinct(sampletype_f) == 3, "Y", "N")) %>% 
  filter(All_Matrices == "Y") %>% 
  ungroup() %>%  
  #Creating plot
  ggplot(aes(x=labid, y= total_pct_recovery)) +
  theme_test() +
  geom_rect(ymin = 95.7, ymax = 104.3, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = sampletype_f, fill = sampletype_f, group = sampletype_f), alpha = 0.7, width = .05, size = 3, position = position_dodge(width = 0.75)) +
  geom_boxplot(alpha = 0.5,aes(color = sampletype_f, fill =sampletype_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darksalmon", "goldenrod4"))+
  scale_color_manual(values = c("royalblue", "darksalmon", "goldenrod4"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Matrix", color = "Matrix", title = "Microscopy", subtitle = "Total Particle Recovery", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(a_a)

b <- Join_Microscopy_Total %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  group_by(sampletype_f, labid, sampleid, total_pct_recovery) %>%
  summarise() %>% 
  rename("Matrix" = sampletype_f) %>% 
  rename("Lab ID" = labid) %>%
  rename("Sample ID" = sampleid) %>%
  rename("Recovery" = total_pct_recovery) %>%
  distinct() 

kable(b, digits = 2, caption = "Microscopy: Total Particle Recovery. Summary statistics by sample (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

#Total - Summary
c<- Join_Microscopy_Total %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, total_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid) %>% 
  mutate(total_pct_recovery_mean = mean(total_pct_recovery)) %>% 
  ungroup() %>% 
  group_by(sampletype_f, labid, microscopy_f, total_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= total_pct_recovery_mean)) +
  theme_test() +
  geom_rect(ymin = 95.7, ymax = 104.3, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = sampletype_f, fill = sampletype_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = sampletype_f, fill = sampletype_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("royalblue", "darksalmon", "goldenrod4")) +   
  scale_fill_manual(values = c("royalblue", "darksalmon", "goldenrod4")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Total Particle Recovery", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(c)

d <- Join_Microscopy_Total %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  group_by(sampletype_f, labid, set) %>% 
  mutate(`Mean (%)`= mean(total_pct_recovery)) %>%  
  mutate(`Median (%)` = median(total_pct_recovery)) %>%
  mutate(`SD (%)` = sd(total_pct_recovery)) %>%
  mutate(`SE (%)` = std.error(total_pct_recovery)) %>% 
  mutate(`Min (%)` = min(total_pct_recovery)) %>% 
  mutate(`Max (%)` = max(total_pct_recovery)) %>%
  ungroup() %>% 
  select(sampletype_f, labid, `Mean (%)`, `Median (%)`, `SD (%)`, `SE (%)`, `Min (%)`, `Max (%)`) %>%
  rename("Matrix" = sampletype_f) %>%  
  rename("Lab ID" = labid) %>%
  distinct() 

kable(d, digits = 2, caption = "Microscopy: Total Particle Recovery. Summary statistics by laboratory (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

e <- Join_Microscopy_Total %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  group_by(sampletype_f, labid, set) %>% 
  mutate(mean_lab = mean(total_pct_recovery)) %>% 
  ungroup() %>%
  distinct(sampletype_f, labid, mean_lab) %>%
  group_by(sampletype_f) %>% 
  mutate(`Mean (%)`= mean(mean_lab)) %>%    
  mutate(`Median (%)` = median(mean_lab)) %>%
  mutate(`SD (%)` = sd(mean_lab)) %>%
  mutate(`SE (%)` = std.error(mean_lab)) %>% 
  mutate(`Min (%)` = min(mean_lab)) %>% 
  mutate(`Max (%)` = max(mean_lab)) %>%
  ungroup() %>% 
  rename("Matrix" = sampletype_f) %>% 
  select(`Matrix`, `Mean (%)`, `Median (%)`, `SD (%)`, `SE (%)`, `Min (%)`, `Max (%)`) %>%
  distinct() 

kable(e, digits = 2, caption = "Microscopy: Total Particle Recovery. Summary statistics across all laboratories. Not blank corrected.")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")
```

### Size Fraction

```{r echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, fig.width = 12, fig.height = 10}

#All Size Fractions
a <- Join_Microscopy_SizeFraction %>% 
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Force size fractions to be in the correct order
  mutate(sizefraction_f = factor(sizefraction_f, levels = c("1-20 µm", "20-212 µm","212-500 µm",">500 µm"))) %>% 
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, sizefraction_f, sizefrac_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, sizefraction_f) %>% 
  mutate(sizefrac_pct_recovery_mean = mean(sizefrac_pct_recovery)) %>% 
  ungroup() %>% 
  group_by(sampletype_f, labid, set, microscopy_f, sizefraction_f, sizefrac_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=sampletype_f, y= sizefrac_pct_recovery_mean)) +
  theme_test() +
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sizefraction_f, ncol = 2, scales = "free")+
  theme(text = element_text(size = 13),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, All Size Fractions", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(a)

#1-20µm - By Lab
b <- Join_Microscopy_SizeFraction %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  filter(sizefraction_f == "1-20 µm") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= sizefrac_pct_recovery)) +
  theme_test() +
  #No RSD data for the 1-20 µm size fraction
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, 1-20 µm", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(b)

#1-20µm - Summary
c <- Join_Microscopy_SizeFraction %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, sizefraction_f, sizefrac_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, sizefraction_f) %>% 
  mutate(sizefrac_pct_recovery_mean = mean(sizefrac_pct_recovery)) %>% 
  ungroup() %>% 
  filter(sizefraction_f == "1-20 µm") %>%
  group_by(sampletype_f, labid, microscopy_f, sizefrac_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= sizefrac_pct_recovery_mean)) +
  theme_test() +
  #No RSD data for the 1-20 µm size fraction
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, 1-20 µm", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(c)

#20-212 µm - By Lab
d <- Join_Microscopy_SizeFraction %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  filter(sizefraction_f == "20-212 µm") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= sizefrac_pct_recovery)) +
  theme_test() +
  geom_rect(ymin = 91.6, ymax = 108.4, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, 20-212 µm", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(d)

#20-212 µm - Summary
e <- Join_Microscopy_SizeFraction %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, sizefraction_f, sizefrac_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, sizefraction_f) %>% 
  mutate(sizefrac_pct_recovery_mean = mean(sizefrac_pct_recovery)) %>% 
  ungroup() %>% 
  filter(sizefraction_f == "20-212 µm") %>%
  group_by(sampletype_f, labid, microscopy_f, sizefrac_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= sizefrac_pct_recovery_mean)) +
  theme_test() +
  geom_rect(ymin = 91.6, ymax = 108.4, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, 20-212 µm", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(e)

#212-500 µm - By Lab
f <- Join_Microscopy_SizeFraction %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  filter(sizefraction_f == "212-500 µm") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= sizefrac_pct_recovery)) +
  theme_test() +
  geom_rect(ymin = 91.05, ymax = 108.95, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, 212-500 µm", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(f)

#212-500 µm - Summary
g <- Join_Microscopy_SizeFraction %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, sizefraction_f, sizefrac_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, sizefraction_f) %>% 
  mutate(sizefrac_pct_recovery_mean = mean(sizefrac_pct_recovery)) %>% 
  ungroup() %>% 
  filter(sizefraction_f == "212-500 µm") %>%
  group_by(sampletype_f, labid, microscopy_f, sizefrac_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= sizefrac_pct_recovery_mean)) +
  theme_test() +
  geom_rect(ymin = 91.05, ymax = 108.95, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, 212-500 µm", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(g)

#>500 µm - By Lab
h <- Join_Microscopy_SizeFraction %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  filter(sizefraction_f == ">500 µm") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= sizefrac_pct_recovery)) +
  theme_test() +
  #No RSD data for the >500 µm size fraction
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, >500 µm", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(h)

#>500 µm - Summary
i <- Join_Microscopy_SizeFraction %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, sizefraction_f, sizefrac_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, sizefraction_f) %>% 
  mutate(sizefrac_pct_recovery_mean = mean(sizefrac_pct_recovery)) %>% 
  ungroup() %>% 
  filter(sizefraction_f == ">500 µm") %>%
  group_by(sampletype_f, labid, microscopy_f, sizefrac_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= sizefrac_pct_recovery_mean)) +
  theme_test() +
  #No RSD data for the >500 µm size fraction
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, >500 µm", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(i)

#Size Fraction - By Sample
j <- Join_Microscopy_SizeFraction %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  group_by(sampletype_f, labid, sampleid, sizefraction_f, sizefrac_pct_recovery) %>% 
  summarise() %>% 
  rename("Matrix" = sampletype_f) %>%
  rename("Lab ID" = labid) %>%
  rename("Sample ID" = sampleid) %>%
  rename("Size Fraction" = sizefraction_f) %>%
  rename("Recovery" = sizefrac_pct_recovery) %>%
  distinct() %>% 
  arrange(`Matrix`, `Lab ID`, `Sample ID`, `Size Fraction`)

kable(j, digits = 2, caption = "Microscopy: Total Particle Recovery by Size Fraction. Summary statistics by sample (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

#Size Fraction - By Lab
k <- Join_Microscopy_SizeFraction %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>% 
  group_by(sampletype_f, labid, set, sizefraction_f) %>% 
  mutate(`Mean (%)`= mean(sizefrac_pct_recovery)) %>%  
  mutate(`Median (%)` = median(sizefrac_pct_recovery)) %>%
  mutate(`SD (%)` = sd(sizefrac_pct_recovery)) %>%
  mutate(`SE (%)` = std.error(sizefrac_pct_recovery)) %>%
  mutate(`Min (%)` = min(sizefrac_pct_recovery)) %>% 
  mutate(`Max (%)` = max(sizefrac_pct_recovery)) %>%
  ungroup() %>% 
  select(sampletype_f, labid, sizefraction_f, `Mean (%)`, `Median (%)`, `SD (%)`, `SE (%)`, `Min (%)`, `Max (%)`) %>%
  rename("Matrix" = sampletype_f) %>%
  rename("Lab ID" = labid) %>%
  rename("Size Fraction" = sizefraction_f) %>%
  distinct() %>% 
  arrange(`Lab ID`, `Size Fraction`)

kable(k, digits = 2, caption = "Microscopy: Total Particle Recovery by Size Fraction. Summary statistics by laboratory (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

#Size Fraction - Summary
l <- k %>% 
  group_by(`Matrix`, `Size Fraction`) %>% 
  mutate(`Global Mean (%)`= mean(`Mean (%)`)) %>%   
  mutate(`Median (%)` = median(`Mean (%)`)) %>%
  mutate(`SD (%)` = sd(`Mean (%)`)) %>%
  mutate(`SE (%)` = std.error(`Mean (%)`)) %>%
  mutate(`Min (%)` = min(`Mean (%)`)) %>% 
  mutate(`Max (%)` = max(`Mean (%)`)) %>%
  ungroup() %>% 
  select(`Matrix`, `Size Fraction`, `Global Mean (%)`, `Median (%)`, `SD (%)`, `SE (%)`, `Min (%)`, `Max (%)`) %>%
  distinct() %>% 
  arrange(`Matrix`,`Size Fraction`)

kable(l, digits = 2, caption = "Microscopy: Total Particle Recovery by Size Fraction. Summary statistics across all laboratories. Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

```

```{r Manuscript Fig S7, echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, include = FALSE}

#All Size Fractions
FigS7 <- Join_Microscopy_SizeFraction %>%  
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  ungroup() %>%
  mutate(sampletype_f = case_when(
    sampletype_f == "Clean Water" ~ "Drinking Water",
    sampletype_f == "Dirty Water" ~ "Surface Water",
    sampletype_f == "Fish Tissue" ~ "Fish Tissue",
    sampletype_f == "Sediment" ~ "Sediment")) %>%
  #Force size fractions to be in the correct order
  mutate(sizefraction_f = factor(sizefraction_f, levels = c("1-20 µm", "20-212 µm","212-500 µm",">500 µm"))) %>%   #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, sizefraction_f, sizefrac_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, sizefraction_f) %>% 
  mutate(sizefrac_pct_recovery_mean = mean(sizefrac_pct_recovery)) %>% 
  ungroup() %>% 
  group_by(sampletype_f, labid, microscopy_f, sizefraction_f, sizefrac_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=sampletype_f, y= sizefrac_pct_recovery_mean)) +
  theme_test() +
  geom_point(aes(color = sampletype_f, fill = sampletype_f), alpha = 0.7, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = sampletype_f, fill = sampletype_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sizefraction_f, nrow = 1, scales = "free")+
  theme(text = element_text(size = 20),
        legend.position = "right",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("royalblue", "darksalmon", "goldenrod4")) +   
  scale_fill_manual(values = c("royalblue", "darksalmon", "goldenrod4")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Matrix", color = "Matrix")

plot(FigS7)

saveRDS(FigS7, file = "./Manuscript_Figures/FigS7")

```

### Color

```{r echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, fig.width = 12, fig.height = 10}

#All Colors
a <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, color_f, color_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, color_f) %>% 
  mutate(color_pct_recovery_mean = mean(color_pct_recovery)) %>% 
  ungroup() %>% 
  group_by(sampletype_f, labid, microscopy_f, color_f, color_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=sampletype_f, y= color_pct_recovery_mean)) +
  theme_test() +
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.7,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~color_f, ncol = 3, scales = "free")+
  theme(text = element_text(size = 13),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", subtitle = "Particle Recovery, All Colors", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(a)

#Blue - By Lab
b <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  filter(color_f == "Blue") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= color_pct_recovery)) +
  theme_test() +
  #No RSD data for blue particles
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, Blue", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(b)

#Blue - Summary
c <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, color_f, color_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, color_f) %>% 
  mutate(color_pct_recovery_mean = mean(color_pct_recovery)) %>% 
  ungroup() %>% 
  filter(color_f == "Blue") %>%
  group_by(sampletype_f, labid, microscopy_f, color_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= color_pct_recovery_mean)) +
  theme_test() +
  #No RSD data for blue particles
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, Blue", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(c)

#White - By Lab
d <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  filter(color_f == "White") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= color_pct_recovery)) +
  theme_test() +
  geom_rect(ymin = 93.5, ymax = 106.5, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, White", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(d)

#White - Summary
e <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, color_f, color_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, color_f) %>% 
  mutate(color_pct_recovery_mean = mean(color_pct_recovery)) %>% 
  ungroup() %>% 
  filter(color_f == "White") %>%
  group_by(sampletype_f, labid, microscopy_f, color_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= color_pct_recovery_mean)) +
  theme_test() +
  geom_rect(ymin = 93.5, ymax = 106.5, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, White", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(e)

#Green - By Lab
f <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  filter(color_f == "Green") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= color_pct_recovery)) +
  theme_test() +
  #No RSD data for green particles
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, Green", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(f)

#Green - Summary
g <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, color_f, color_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, color_f) %>% 
  mutate(color_pct_recovery_mean = mean(color_pct_recovery)) %>% 
  ungroup() %>% 
  filter(color_f == "Green") %>%
  group_by(sampletype_f, labid, microscopy_f, color_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= color_pct_recovery_mean)) +
  theme_test() +
  #No RSD data for green particles
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, Green", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(g)

#Orange - By Lab
h <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  filter(color_f == "Orange") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= color_pct_recovery)) +
  theme_test() +
  #No RSD data for orange particles
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, Orange", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(h)

#Orange - Summary
i <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, color_f, color_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, color_f) %>% 
  mutate(color_pct_recovery_mean = mean(color_pct_recovery)) %>% 
  ungroup() %>% 
  filter(color_f == "Orange") %>%
  group_by(sampletype_f, labid, microscopy_f, color_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= color_pct_recovery_mean)) +
  theme_test() +
  #No RSD data for orange particles
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, Orange", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(i)

#Clear - By Lab
j <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  filter(color_f == "Clear") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= color_pct_recovery)) +
  theme_test() +
  #No RSD data for Clear particles
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, Clear", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(j)

#Clear - Summary
k <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, color_f, color_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, color_f) %>% 
  mutate(color_pct_recovery_mean = mean(color_pct_recovery)) %>% 
  ungroup() %>% 
  filter(color_f == "Clear") %>%
  group_by(sampletype_f, labid, microscopy_f, color_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= color_pct_recovery_mean)) +
  theme_test() +
  #No RSD data for Clear particles
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, Clear", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(k)

#Color - By Sample
l <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  group_by(sampletype_f, labid, sampleid, color_f, color_pct_recovery) %>% 
  summarise() %>% 
  rename("Matrix" = sampletype_f) %>%
  rename("Lab ID" = labid) %>%
  rename("Sample ID" = sampleid) %>%
  rename("Color" = color_f) %>%
  rename("Recovery" = color_pct_recovery) %>%
  distinct() 

kable(l, digits = 2, caption = "Microscopy: Total Particle Recovery by Size Fraction. Summary statistics by sample (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

#Color - By Lab
m <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  group_by(sampletype_f, labid, color_f, set) %>% 
  mutate(`Mean (%)`= mean(color_pct_recovery)) %>%  
  mutate(`Median (%)` = median(color_pct_recovery)) %>%
  mutate(`SD (%)` = sd(color_pct_recovery)) %>%
  mutate(`SE (%)` = std.error(color_pct_recovery)) %>%
  mutate(`Min (%)` = min(color_pct_recovery)) %>% 
  mutate(`Max (%)` = max(color_pct_recovery)) %>%
  ungroup() %>% 
  select(sampletype_f, labid, color_f, `Mean (%)`, `Median (%)`, `SD (%)`, `SE (%)`, `Min (%)`, `Max (%)`) %>%
  rename("Matrix" = sampletype_f) %>%
  rename("Lab ID" = labid) %>%
  rename("Color" = color_f) %>%
  distinct() 

kable(m, digits = 2, caption = "Microscopy: Total Particle Recovery by Color. Summary statistics by laboratory (n = 1-3 samples). Not blank corrected.")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

#Color - Summary
n <- m %>%
  group_by(`Matrix`, `Color`) %>% 
  mutate(`Global Mean (%)`= mean(`Mean (%)`)) %>%   
  mutate(`Median (%)` = median(`Mean (%)`)) %>%
  mutate(`SD (%)` = sd(`Mean (%)`)) %>%
  mutate(`SE (%)` = std.error(`Mean (%)`)) %>%
  mutate(`Min (%)` = min(`Mean (%)`)) %>% 
  mutate(`Max (%)` = max(`Mean (%)`)) %>%
  ungroup() %>% 
  select(`Matrix`, `Color`, `Global Mean (%)`, `Median (%)`, `SD (%)`, `SE (%)`, `Min (%)`, `Max (%)`) %>%
  distinct() 

kable(n, digits = 2, caption = "Microscopy: Total Particle Recovery by Color. Summary statistics across all laboratories. Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")
  
```

```{r Manuscript Fig S8, echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, include = FALSE}

#All Colors
FigS8 <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  mutate(sampletype_f = case_when(
    sampletype_f == "Clean Water" ~ "Drinking Water",
    sampletype_f == "Dirty Water" ~ "Surface Water",
    sampletype_f == "Fish Tissue" ~ "Fish Tissue",
    sampletype_f == "Sediment" ~ "Sediment")) %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, color_f, color_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, color_f) %>% 
  mutate(color_pct_recovery_mean = mean(color_pct_recovery)) %>% 
  ungroup() %>% 
  group_by(sampletype_f, labid, microscopy_f, color_f, color_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=sampletype_f, y= color_pct_recovery_mean)) +
  theme_test() +
  geom_point(aes(color = sampletype_f, fill = sampletype_f), alpha = 0.7, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = sampletype_f, fill = sampletype_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~color_f, nrow = 1, scales = "free")+
  theme(text = element_text(size = 20),
        legend.position = "right",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("royalblue", "darksalmon", "goldenrod4")) +   
  scale_fill_manual(values = c("royalblue", "darksalmon", "goldenrod4")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Matrix", color = "Matrix") 

plot(FigS8)

saveRDS(FigS8, file = "./Manuscript_Figures/FigS8")

```

### Morphology

```{r echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, fig.width = 12, fig.height = 10}

#All Morphologies
a <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, morphology_f, morphology_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, morphology_f) %>% 
  mutate(morphology_pct_recovery_mean = mean(morphology_pct_recovery)) %>% 
  ungroup() %>% 
  group_by(sampletype_f, labid, microscopy_f, morphology_f, morphology_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=sampletype_f, y= morphology_pct_recovery_mean)) +
  theme_test() +
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.7,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~morphology_f, ncol = 3, scales = "free")+
  theme(text = element_text(size = 13),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, All Morphologies", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(a)


#Fibers - By Lab
b <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  filter(morphology_f == "Fiber") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= morphology_pct_recovery)) +
  theme_test() +
  #No RSD data for fibers
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, Fibers", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(b)

#Fiber - Summary
c <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, morphology_f, morphology_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, morphology_f) %>% 
  mutate(morphology_pct_recovery_mean = mean(morphology_pct_recovery)) %>% 
  ungroup() %>% 
  filter(morphology_f == "Fiber") %>%
  group_by(sampletype_f, labid, microscopy_f,morphology_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= morphology_pct_recovery_mean)) +
  theme_test() +
  #No RSD data for fibers
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, Fibers", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(c)

#Spheres - By Lab
d <- Join_Microscopy_Morphology %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  filter(morphology_f == "Sphere") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= morphology_pct_recovery)) +
  theme_test() +
  geom_rect(ymin = 98.2, ymax = 101.8, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, Spheres", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(d)

#Sphere - Summary
e <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, morphology_f, morphology_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, morphology_f) %>% 
  mutate(morphology_pct_recovery_mean = mean(morphology_pct_recovery)) %>% 
  ungroup() %>% 
  filter(morphology_f == "Sphere") %>%
  group_by(sampletype_f, labid, microscopy_f,morphology_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= morphology_pct_recovery_mean)) +
  theme_test() +
  geom_rect(ymin = 98.2, ymax = 101.8, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, Sphere", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(e)

#Fragments - By Lab
f <- Join_Microscopy_Morphology %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  filter(morphology_f == "Fragment") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= morphology_pct_recovery)) +
  theme_test() +
  geom_rect(ymin = 93.6, ymax = 106.4, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, Fragments", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(f)

#Fragments - Summary
g <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, morphology_f, morphology_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, morphology_f) %>% 
  mutate(morphology_pct_recovery_mean = mean(morphology_pct_recovery)) %>% 
  ungroup() %>% 
  filter(morphology_f == "Fragment") %>%
  group_by(sampletype_f, labid, microscopy_f,morphology_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= morphology_pct_recovery_mean)) +
  theme_test() +
  geom_rect(ymin = 93.6, ymax = 106.4, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +facet_wrap(~sampletype_f, ncol = 3)+
  geom_hline(yintercept = 100, linetype = "dotted")+
  
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, Fragments", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(g)

#Morphology - By Sample
h <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  group_by(sampletype_f, labid, sampleid, morphology_f, morphology_pct_recovery) %>% 
  summarise() %>%
  rename("Matrix" = sampletype_f) %>% 
  rename("Lab ID" = labid) %>%
  rename("Sample ID" = sampleid) %>%
  rename("Morphology" = morphology_f) %>%
  rename("Recovery" = morphology_pct_recovery) %>%
  distinct() 

kable(h, digits = 2, caption = "Microscopy: Total Particle Recovery by Size Fraction. Summary statistics by sample (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

#Morphology - By Lab
i <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  group_by(sampletype_f, labid, set, morphology_f) %>% 
  mutate(`Mean (%)`= mean(morphology_pct_recovery)) %>%  
  mutate(`Median (%)` = median(morphology_pct_recovery)) %>%
  mutate(`SD (%)` = sd(morphology_pct_recovery)) %>%
  mutate(`SE (%)` = std.error(morphology_pct_recovery)) %>%
  mutate(`Min (%)` = min(morphology_pct_recovery)) %>% 
  mutate(`Max (%)` = max(morphology_pct_recovery)) %>%
  ungroup() %>% 
  select(sampletype_f, labid, morphology_f, `Mean (%)`, `Median (%)`, `SD (%)`, `SE (%)`, `Min (%)`, `Max (%)`) %>%
  rename("Matrix" = sampletype_f) %>% 
  rename("Lab ID" = labid) %>%
  rename("Morphology" = morphology_f) %>%
  distinct() 

kable(i, digits = 2, caption = "Microscopy: Total Particle Recovery by Morphology. Summary statistics by laboratory (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px") 

#Morphology - Summary
j <- Join_Microscopy_Morphology %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  group_by(sampletype_f, morphology_f) %>% 
  mutate(`Mean (%)`= mean(morphology_pct_recovery)) %>%  
  mutate(`Median (%)` = median(morphology_pct_recovery)) %>%
  mutate(`SD (%)` = sd(morphology_pct_recovery)) %>%
  mutate(`SE (%)` = std.error(morphology_pct_recovery)) %>%
  mutate(`Min (%)` = min(morphology_pct_recovery)) %>% 
  mutate(`Max (%)` = max(morphology_pct_recovery)) %>%
  ungroup() %>% 
  select(sampletype_f, morphology_f, `Mean (%)`, `Median (%)`, `SD (%)`, `SE (%)`, `Min (%)`, `Max (%)`) %>%
  rename("Matrix" = sampletype_f) %>%
  rename("Morphology" = morphology_f) %>%
  distinct() 

kable(j, digits = 2, caption = "Microscopy: Total Particle Recovery by Morphology. Summary statistics across all laboratories. Not blank corrected.")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")
```

```{r Manuscript Fig S9, echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, include = FALSE}

FigS9 <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  mutate(sampletype_f = case_when(
    sampletype_f == "Clean Water" ~ "Drinking Water",
    sampletype_f == "Dirty Water" ~ "Surface Water",
    sampletype_f == "Fish Tissue" ~ "Fish Tissue",
    sampletype_f == "Sediment" ~ "Sediment")) %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, morphology_f, morphology_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, morphology_f) %>% 
  mutate(morphology_pct_recovery_mean = mean(morphology_pct_recovery)) %>% 
  ungroup() %>% 
  group_by(sampletype_f, labid, microscopy_f, morphology_f, morphology_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=sampletype_f, y= morphology_pct_recovery_mean)) +
  theme_test() +
  geom_point(aes(color = sampletype_f, fill = sampletype_f), alpha = 0.7, size = 3) +
  geom_boxplot(alpha = 0.7,aes(color = sampletype_f, fill = sampletype_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~morphology_f, ncol = 3, scales = "free")+
  theme(text = element_text(size = 20),
        legend.position = "right",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("royalblue", "darksalmon", "goldenrod4")) +   
  scale_fill_manual(values = c("royalblue", "darksalmon", "goldenrod4")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Matrix", color = "Matrix") 

plot(FigS9)

saveRDS(FigS9, file = "./Manuscript_Figures/FigS9")

```

### Color and Morphology within Size Fraction

The following plots display the recovery (reported count/spiked count (%)) obtained from microscopy. **Reported particle characteristics must match expected particle characteristics of the spiking materials to count towards recovery estimations.** For example, the colors of polymer spiking materials are defined as white, orange, blue, clear and green. Colors outside of these definitions are currently excluded from analysis. 

The dotted line is at 100% and the grey area represents the mean RSD for each given category. The data are **not** blank corrected.

Recovery in this section is estimated based on color and morphology within a size fraction. For example, if the sample was spiked with 10 green spheres >500 µm and 8 green spheres were recovered in the >500 µm size fraction, recovery is estimated as 80%. However, if 4 of these spheres were reported as blue rather than green, recovery would fall to 60%. 

```{r echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, results = 'asis'}

#Make data frame of particle characteristics
df <- Join_Microscopy_Combo %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f == "Y") %>% 
  group_by(sampletype_f, morphology_f, color_f, sizefraction_f) %>% 
  summarise()

for (i in 1:nrow(df))

{ #Open loop  
  
#By Lab
a <- Join_Microscopy_Combo %>%
  group_by(sampletype_f, labid, sampleid, microscopy_f,sizefraction_f, color_f, morphology_f, color_morphology_sizefraction_pct_recovery) %>% 
  filter(sizefraction_f == df$sizefraction_f[[i]]) %>% 
  filter(color_f == df$color_f[[i]]) %>% 
  filter(morphology_f == df$morphology_f[[i]]) %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%  
  #Creating plot
  ggplot(aes(x=labid, y= color_morphology_sizefraction_pct_recovery)) +
  theme_test() +
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  facet_wrap(~sampletype_f, scales = "free")+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", 
       subtitle = paste(df$sizefraction_f[[i]], df$color_f[[i]], df$morphology_f[[i]]), 
       caption = "Points represent individual samples spiked with microplastic particles.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(a)

#Summary
b <- Join_Microscopy_Combo %>%
  filter(sizefraction_f == df$sizefraction_f[[i]]) %>% 
  filter(color_f == df$color_f[[i]]) %>% 
  filter(morphology_f == df$morphology_f[[i]]) %>%
  group_by(sampletype_f, labid) %>%
  mutate(color_morphology_sizefraction_pct_recovery_mean = mean(color_morphology_sizefraction_pct_recovery)) %>%
  ungroup() %>%
  group_by(sampletype_f, labid, microscopy_f,color_morphology_sizefraction_pct_recovery_mean) %>%
  summarise() %>%
  #Creating plot
  ggplot(aes(x=microscopy_f,y= color_morphology_sizefraction_pct_recovery_mean)) +
  theme_test() +
  geom_point(aes(color = microscopy_f, fill = microscopy_f), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f,fill = microscopy_f), outlier.shape = NA) +
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  facet_wrap(~sampletype_f, scales = "free")+
  scale_color_manual(values = c("darkmagenta", "orchid")) +
  scale_fill_manual(values = c("darkmagenta", "orchid")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Method", color = "Method", title = "Microscopy", 
       subtitle = paste(df$sizefraction_f[[i]], df$color_f[[i]], df$morphology_f[[i]]),
       caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(b)
 
#By Sample
c <- Join_Microscopy_Combo %>%
  group_by(sampletype_f, labid, sizefraction_f, color_f, morphology_f) %>%
  filter(sizefraction_f == df$sizefraction_f[[i]]) %>% 
  filter(color_f == df$color_f[[i]]) %>% 
  filter(morphology_f == df$morphology_f[[i]]) %>%
  ungroup() %>%
  select(sampletype_f, labid, sampleid, color_morphology_sizefraction_pct_recovery) %>%
  rename("Matrix" = sampletype_f) %>% 
  rename("Lab ID" = labid) %>%
  rename("Sample ID" = sampleid) %>%
  rename("Recovery" = color_morphology_sizefraction_pct_recovery) %>%
  distinct()

print(kable(c, digits = 2, caption = paste(df$sizefraction_f[[i]], df$color_f[[i]], df$morphology_f[[i]])) %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px"))
  cat('\n')

#By Lab
d <- Join_Microscopy_Combo %>%
  group_by(sampletype_f, labid, sizefraction_f, color_f, morphology_f) %>%
  filter(sizefraction_f == df$sizefraction_f[[i]]) %>% 
  filter(color_f == df$color_f[[i]]) %>% 
  filter(morphology_f == df$morphology_f[[i]]) %>%
  mutate(Mean = mean(color_morphology_sizefraction_pct_recovery)) %>%
  mutate(Median = median(color_morphology_sizefraction_pct_recovery)) %>%
  mutate(SD = sd(color_morphology_sizefraction_pct_recovery)) %>%
  mutate(SE = std.error(color_morphology_sizefraction_pct_recovery)) %>%
  mutate(Min = min(color_morphology_sizefraction_pct_recovery)) %>%
  mutate(Max = max(color_morphology_sizefraction_pct_recovery)) %>%
  ungroup() %>%
  select(sampletype_f, labid, Mean, Median, SD, SE, Min, Max) %>%
  rename("Matrix" = sampletype_f) %>% 
  rename("Lab ID" = labid) %>%
  rename("Mean (%)" = Mean) %>%
  rename("Median (%)" = Median) %>%
  rename("SD (%)" = SD) %>%
  rename("SE (%)" = SE) %>%
  rename("Min (%)" = Min) %>%
  rename("Max (%)" = Max) %>%
  distinct()

print(kable(d, digits = 2, caption = paste(df$sizefraction_f[[i]], df$color_f[[i]], df$morphology_f[[i]])) %>% 
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px"))  
  cat('\n')

 } #Close loop

```

### Team Recov stuff

```{r echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, fig.width = 12, fig.height = 10, fig.height = 10}
#Total - Summary
c<- Join_Microscopy_Total %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  filter(!(labid %in% c("AA", "L", "M", "Q", "R"))) %>%
  #Data manipulation
  group_by(sampletype_f, labid, sampleid, microscopy_f, total_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid) %>% 
  mutate(total_pct_recovery_mean = mean(total_pct_recovery)) %>% 
  ungroup() %>% 
  group_by(sampletype_f, labid, microscopy_f, total_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= total_pct_recovery_mean)) +
  theme_test() +
  geom_rect(ymin = 95.7, ymax = 104.3, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = sampletype_f, fill = sampletype_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = sampletype_f, fill = sampletype_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 3)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("royalblue", "darksalmon", "goldenrod4")) +   
  scale_fill_manual(values = c("royalblue", "darksalmon", "goldenrod4")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Total Particle Recovery", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(c)

e <- Join_Microscopy_Total %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  filter(!(labid %in% c("AA", "L", "M", "Q", "R"))) %>%
  group_by(sampletype_f, labid) %>% 
  mutate(mean_lab = mean(total_pct_recovery)) %>% 
  ungroup() %>%
  distinct(sampletype_f, labid, mean_lab) %>%
  group_by(sampletype_f) %>% 
  mutate(`Mean (%)`= mean(mean_lab)) %>%    
  mutate(`Median (%)` = median(mean_lab)) %>%
  mutate(`SD (%)` = sd(mean_lab)) %>%
  mutate(`SE (%)` = std.error(mean_lab)) %>%
  mutate(`Min (%)` = min(mean_lab)) %>% 
  mutate(`Max (%)` = max(mean_lab)) %>%
  ungroup() %>% 
  rename("Matrix" = sampletype_f) %>% 
  select(`Matrix`, `Mean (%)`, `Median (%)`, `SD (%)`, `SE (%)`, `Min (%)`, `Max (%)`) %>%
  distinct() 

kable(e, digits = 2, caption = "Microscopy: Total Particle Recovery. Summary statistics across all laboratories. Not blank corrected.")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) #%>% 
  # scroll_box(width = "700px", height = "200px")

d <- Join_Microscopy_Total %>%
  filter(sampletype_f != "Dirty Water") %>%
  filter(plastic_f != "N") %>%
  filter(!(labid %in% c("AA", "L", "M", "Q", "R"))) %>%
  group_by(sampletype_f, labid) %>% 
  mutate(`Mean (%)`= mean(total_pct_recovery)) %>%  
  mutate(`Median (%)` = median(total_pct_recovery)) %>%
  mutate(`SD (%)` = sd(total_pct_recovery)) %>%
  mutate(`SE (%)` = std.error(total_pct_recovery)) %>%
  mutate(`Min (%)` = min(total_pct_recovery)) %>% 
  mutate(`Max (%)` = max(total_pct_recovery)) %>%
  ungroup() %>% 
  select(sampletype_f, labid, `Mean (%)`, `Median (%)`, `SD (%)`, `SE (%)`, `Min (%)`, `Max (%)`) %>%
  rename("Matrix" = sampletype_f) %>%  
  rename("Lab ID" = labid) %>%
  distinct() 

kable(d, digits = 2, caption = "Microscopy: Total Particle Recovery. Summary statistics by laboratory (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")


```

# Clean Water, Fish Tissue, Sediment & Dirty Water Comparisons > 212 µm

The following analyses are identical to those above, but spiked particles that are predicted to fall into size fractions less than 212 µm or particles that were found in size fractions less than 212 µm are excluded from analysis so that comparisons may be made across all matrices. More specifically, only the following size fractions are included: 212-500 µm, 250-500 µm and >500 µm. 

```{r, echo = FALSE, results = 'hide', error = FALSE, warning = FALSE, message = FALSE}

#WARNING: VARIABLE NAMES ARE THE SAME AS ABOVE. 
#THEY WILL BE WRITTEN OVER IN THE ENVIRONMENT. 
#THIS SCRIPT PICKS UP AT THE BEGINNING OF MICROSCOPY DATA TIDYING 

# Results are selected to only include particles that were identified via microscopy. Counts for unique samples and particle IDs for total, size fraction, color and morphology are calculated and a new column is created for each.

#Calculate the number of particles for each characteristic and combination 

Results_Clean_Microscopy <- Results_Clean %>%
  
  #Size fraction filter
  filter(sizefraction_f %in% c("212-500 µm", "250-500 µm", ">500 µm", ">212 µm")) %>% 
  
  #Select particles analyzed by microscopy
  filter(microscopy_f == "Yes") %>%
  group_by(sampletype_f, sampleid) %>% 
  #Total particles per sample
  mutate(total = n_distinct(sampleid, particleid)) %>% 
  ungroup() %>% 
  #Total particles by size fraction
  group_by(sampletype_f, sampleid, sizefraction_f) %>% 
  mutate(total_sizefraction = n_distinct(sampleid, particleid)) %>%  
  ungroup() %>% 
  #Total particles by color
  group_by(sampletype_f, sampleid, color_f) %>% 
  mutate(total_color = n_distinct(sampleid, particleid)) %>% 
  ungroup() %>% 
  #Total particles by morphology
  group_by(sampletype_f, sampleid, morphology_f) %>% 
  mutate(total_morphology = n_distinct(sampleid, particleid)) %>% 
  ungroup() %>% 
  #Total particles by color and morphology within size fraction
  group_by(sampletype_f, sampleid, sizefraction_f, color_f, morphology_f) %>% 
  mutate(total_color_morphology_sizefraction = n_distinct(sampleid, particleid)) %>% 
  ungroup() %>% 
  #Total particles by color and morphology
  group_by(sampletype_f, sampleid, color_f, morphology_f) %>% 
  mutate(total_color_morphology = n_distinct(sampleid, particleid)) %>% 
  ungroup() %>% 
  #Select and rename relevant type of expertise to work for fields below
  rename("expertise" = expertisevisualmicroscopy) %>% 
  replace_na(list(expertise = "Not Reported")) 

```

```{r, echo = FALSE, results = 'hide', error = FALSE, warning = FALSE, message = FALSE}

#Particle counts within each sample are totaled for the following categories: total, size fraction, morphology, color, and polymer. A new column is created for each categories. Values are sample specific. 

Reference_Clean <- Reference %>%
  #Remove small size fraction spiking data from lab R that isn't in other samples
  filter(!(labid == "R" & sizefraction == "20-Jan" & sampletype == "DW")) %>% 
  filter(!(labid == "R" & sizefraction == "20-212" & sampletype == "DW")) %>% 
  
  #Drop spiking values of 0 or NAs
  filter(!averagenumber == 0) %>% 
  
  #Create factor for color
  mutate(color = case_when(
    color == "white" ~ "White",
    color == "orange" ~ "Orange",
    color == "green" ~ "Green",
    color == "red" ~ "Red",
    color == "blue" ~ "Blue",
    color == "clear" ~ "Clear")) %>% 
  mutate(color_f = as.factor(color)) %>% 
  #Create factor for shape
  mutate(morphology = case_when(
    morphology == "fragment" ~ "Fragment",
    morphology == "sphere" ~ "Sphere",
    morphology == "fiber" ~ "Fiber")) %>%  
    mutate(morphology_f = as.factor(morphology)) %>% 
  #Create factor for size fraction
   mutate(sizefraction = case_when(
    sizefraction == "20-Jan" ~ "1-20 µm",
    sizefraction == "1-20" ~ "1-20 µm",
    sizefraction == "20-212" ~ "20-212 µm",
    sizefraction == "212-500" ~ "212-500 µm",
    sizefraction == ">500" ~ ">500 µm")) %>%   
  mutate(sizefraction_f = as.factor(sizefraction)) %>%
  mutate(plastic_f = as.factor(plastic)) %>%
  mutate(sampletype = case_when(
    sampletype == "CW" ~ "Clean Water",
    sampletype == "FT" ~ "Fish Tissue",
    sampletype == "SD" ~ "Sediment",
    sampletype == "DW" ~ "Dirty Water")) %>% 
  mutate(sampletype_f = as.factor(sampletype)) %>%
  
  #Size fraction filter
  filter(sizefraction_f %in% c("212-500 µm", "250-500 µm", ">500 µm")) %>% 
  
#All of the following calculations distinguish between plastic and non-plastic spikes
  #Total particle spike per sample
  group_by(sampletype_f, sampleid, plastic_f) %>% 
  mutate(total_spike = sum(averagenumber, na.rm = TRUE)) %>% 
  ungroup() %>% 
  #Total particle spike per size fraction
  group_by(sampletype_f, sampleid, plastic_f, sizefraction_f) %>% 
  mutate(total_spike_sizefraction = sum(averagenumber, na.rm = TRUE)) %>% 
  ungroup() %>%
  #Total particle spike per color
  group_by(sampletype_f, sampleid, plastic_f, color_f) %>% 
  mutate(total_spike_color = sum(averagenumber, na.rm = TRUE)) %>%  
  ungroup() %>%
  #Total particle spike per morphology
  group_by(sampletype_f, sampleid, plastic_f, morphology_f) %>% 
  mutate(total_spike_morphology = sum(averagenumber, na.rm = TRUE)) %>% 
  ungroup()%>%
  #Total particle spike per size fraction, color and morphology
  group_by(sampletype_f, sampleid, plastic_f, sizefraction_f, color_f, morphology_f) %>% 
  mutate(total_spike_color_morphology_sizefraction = sum(averagenumber, na.rm = TRUE)) %>% 
  ungroup() %>% 
  #Total particle spike per color and morphology
  group_by(sampletype_f, sampleid, plastic_f, color_f, morphology_f) %>% 
  mutate(total_spike_color_morphology = sum(averagenumber, na.rm = TRUE)) %>% 
  ungroup() 

```

```{r, echo = FALSE, error = FALSE, warning = FALSE, message = FALSE}

#Join relevant columns of reference data frame and results to compare expected counts to actual counts for microscopy only data. Recovery for each category is calculated on a sample-specific basis. A new recovery column is created for each category. 

#Total
totalspikecol <- Reference_Clean %>% #Select total_spike column from reference
  distinct(sampletype_f, sampleid, plastic_f, total_spike)
  
Join_Microscopy_Total <- inner_join(Results_Clean_Microscopy, totalspikecol, by = c("sampleid", "sampletype_f")) %>% #Join total_spike column to microscopy results data frame
    mutate(total_pct_recovery = ((total/total_spike)*100)) %>%  #Create a new column for the total recovery of microscopy only results
    group_by(labid, set, sampletype_f, sampleid, plastic_f, microscopy_f, total, total_spike, total_pct_recovery, expertise) %>% 
    summarise()
    
#Size Fraction
sizefractionspikecol <- Reference_Clean %>% 
  distinct(sampletype_f, sampleid, plastic_f, sizefraction_f, total_spike_sizefraction)
  
sizefractionsumresults <- Results_Clean_Microscopy %>% 
  filter(!grepl("4", sampleid)) %>%
  filter(!grepl("8", sampleid)) %>% #Dropping blank samples from results
  drop_na(sizefraction_f) %>% #Drop particles not assigned to a size fraction
  group_by(labid, set, sampletype_f, sampleid, microscopy_f, sizefraction_f, total_sizefraction, expertise) %>% 
  summarise()

Join_Microscopy_SizeFraction <- inner_join(sizefractionsumresults, sizefractionspikecol, by = c("sampleid" = "sampleid", "sizefraction_f" = "sizefraction_f", "sampletype_f")) %>%  
    mutate(sizefrac_pct_recovery = ((total_sizefraction/total_spike_sizefraction)*100)) 

#Color
colorspikecol <- Reference_Clean %>% 
  distinct(sampletype_f, sampleid, plastic_f, color_f, total_spike_color)
  
colorsumresults <- Results_Clean_Microscopy %>% 
  filter(!grepl("4", sampleid)) %>%
  filter(!grepl("8", sampleid)) %>% #Dropping blank samples from results
  group_by(labid, set, sampletype_f, sampleid, microscopy_f, color_f, total_color, expertise) %>% 
  summarise()

Join_Microscopy_Color <- inner_join(colorsumresults, colorspikecol, by = c("sampleid" = "sampleid", "color_f" = "color_f", "sampletype_f")) %>%  
    mutate(color_pct_recovery = ((total_color/total_spike_color)*100)) 

#Morphology
morphologyspikecol <- Reference_Clean %>% 
  distinct(sampletype_f, sampleid, plastic_f, morphology_f, total_spike_morphology)
  
morphologysumresults <- Results_Clean_Microscopy %>% 
  filter(!grepl("4", sampleid)) %>%
  filter(!grepl("8", sampleid)) %>% #Dropping blank samples from results
  group_by(labid, set, sampletype_f, sampleid, microscopy_f, morphology_f, total_morphology, expertise) %>% 
  summarise()

Join_Microscopy_Morphology <- inner_join(morphologysumresults, morphologyspikecol, by = c("sampleid" = "sampleid", "morphology_f" = "morphology_f", "sampletype_f")) %>%  
    mutate(morphology_pct_recovery = ((total_morphology/total_spike_morphology)*100))

#Combination - Color and Morphology Only

 combocmspikecol <- Reference_Clean %>%
   distinct(sampletype_f, sampleid, plastic_f, color_f, morphology_f, total_spike_color_morphology)

 combosumresultscm <- Results_Clean_Microscopy %>%
   filter(!grepl("4", sampleid)) %>%
   filter(!grepl("8", sampleid)) %>% #Dropping blank samples from results
   group_by(labid, set, sampletype_f, sampleid, microscopy_f, color_f, morphology_f, total_color_morphology, expertise) %>%
   summarise()

 Join_Microscopy_Combo_CM <- inner_join(combosumresultscm, combocmspikecol, by = c("sampleid" = "sampleid", "morphology_f" = "morphology_f", "color_f" = "color_f", "sampletype_f")) %>%
   mutate(color_morphology_pct_recovery = ((total_color_morphology/total_spike_color_morphology)*100))


#Combination - Size Fraction, Color and Morphology 

 combospikecol <- Reference_Clean %>%
   distinct(sampletype_f, sampleid, plastic_f, sizefraction_f, color_f, morphology_f, total_spike_color_morphology_sizefraction)

 combosumresults <- Results_Clean_Microscopy %>%
   filter(!grepl("4", sampleid)) %>%
   filter(!grepl("8", sampleid)) %>% #Dropping blank samples from results
   drop_na(sizefraction_f) %>% #Drop particles not assigned to a size fraction
   group_by(labid, set, sampletype_f, sampleid, microscopy_f, sizefraction_f, color_f, morphology_f, total_color_morphology_sizefraction, expertise) %>%
   summarise()

 Join_Microscopy_Combo <- inner_join(combosumresults, combospikecol, by = c("sampleid" = "sampleid", "morphology_f" = "morphology_f", "color_f" = "color_f", "sizefraction_f" = "sizefraction_f", "sampletype_f")) %>%
   mutate(color_morphology_sizefraction_pct_recovery = ((total_color_morphology_sizefraction/total_spike_color_morphology_sizefraction)*100))

```

## Counts

The following section describes data as absolute particle counts that were reported in each sample. Note that Dirty Water samples were not spiked with any particles less than 212µm and are therefore distinct from the other matrices.  

```{r echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, fig.width = 12, fig.height = 10}

spike <- Join_Microscopy_Total %>%
  #Remove spiked particles with problematic RSD
  filter(!(labid == "R" & sampletype_f == "Clean Water")) %>% 
  filter(plastic_f != "N") %>% 
  ungroup() %>%
  group_by(sampletype_f) %>% 
  mutate(mean_spike = mean(total_spike)) 

#Total - By Lab

a <- Results_Clean_Microscopy %>% 
  #Data manipulation
  filter(!grepl("4", sampleid)) %>%
  filter(!grepl("8", sampleid)) %>% #Dropping blanks
  group_by(labid, sampleid, total, expertise, sampletype_f) %>% 
  summarise() %>% 
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%   
  #Creating plot
  ggplot(aes(x= labid, y= total)) +
  theme_test() +
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(data = spike, aes(yintercept = mean_spike), linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  facet_wrap(~sampletype_f, ncol = 2)+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Particle Count", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Raw Particle Count", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line represents the number of spiked polymer particles.\nNot blank corrected.")

plot(a)

b <- Results_Clean_Microscopy %>% 
  filter(blank != "Y") %>% 
  select(sampletype_f, labid, sampleid, total) %>%
  rename("Matrix" = sampletype_f) %>%
  rename("Lab ID" = labid) %>%
  rename("Sample ID" = sampleid) %>%
  rename("Total Particles" = total) %>%
  distinct() 

kable(b, digits = 2, caption = "Raw Particle Count. Summary statistics by sample (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

#Total - Summary
c <- Results_Clean_Microscopy %>%  
  filter(!grepl("4", sampleid)) %>%
  filter(!grepl("8", sampleid)) %>% #Dropping blanks
  group_by(labid, set, sampleid, microscopy_f, total, sampletype_f) %>% 
  summarise() %>% 
  group_by(labid) %>% 
  mutate(total_mean = mean(total)) %>% 
  ungroup() %>% 
  group_by(labid, microscopy_f, total_mean, sampletype_f) %>% 
  summarise() %>%   
  #Creating plot
  ggplot(aes(x=microscopy_f, y= total_mean)) +
  theme_test() +
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_hline(data = spike, aes(yintercept = mean_spike), linetype = "dotted")+
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +
  scale_fill_manual(values = c("darkmagenta")) +
  facet_wrap(~sampletype_f, ncol = 2)+
  expand_limits(y = 0) +
  labs(y = "Particle Count", fill = "Lab", title = "Microscopy", subtitle = "Raw Particle Count", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents the number of spiked polymer particles.\nNot blank corrected.")
  
plot(c)

d <- Results_Clean_Microscopy %>% 
  filter(blank != "Y") %>% 
  group_by(sampletype_f, labid, set, total) %>% 
  summarise() %>% 
  mutate(`Mean`= mean(total)) %>%  
  mutate(`Median` = median(total)) %>%
  mutate(`SD` = sd(total)) %>%
  mutate(`Min` = min(total)) %>% 
  mutate(`Max` = max(total)) %>%
  ungroup() %>% 
  select(sampletype_f, labid, `Mean`, `Median`, `SD`, `Min`, `Max`) %>%
  rename("Matrix" = sampletype_f) %>%
  rename("Lab ID" = labid) %>%
  distinct() 

kable(d, digits = 2, caption = "Raw Particle Count. Summary statistics by laboratory (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px")

e <- Results_Clean_Microscopy %>%  
  filter(!grepl("4", sampleid)) %>%
  filter(!grepl("8", sampleid)) %>% #Dropping blanks
  group_by(sampletype_f, labid, set, total) %>% 
  summarise() %>% 
  mutate(total_mean = mean(total)) %>% 
  ungroup() %>% 
  distinct(sampletype_f, labid, total_mean) %>%
  group_by(sampletype_f) %>% 
  mutate(`Mean`= mean(total_mean)) %>%  
  mutate(`Median` = median(total_mean)) %>%
  mutate(`SD` = sd(total_mean)) %>%
  mutate(`Min` = min(total_mean)) %>% 
  mutate(`Max` = max(total_mean)) %>%
  ungroup() %>% 
  select(sampletype_f, `Mean`, `Median`, `SD`, `Min`, `Max`) %>%
  rename("Matrix" = sampletype_f) %>%
  distinct() 

kable(e, digits = 2, caption = "Raw Particle Count. Global summary statistics. Not blank corrected.")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

#Size Fraction Count
f <- Results_Clean_Microscopy %>%  
  filter(!grepl("4", sampleid)) %>%
  filter(!grepl("8", sampleid)) %>% #Dropping blanks
  group_by(sampletype_f, sizefraction_f, labid, set, total_sizefraction) %>% 
  summarise() %>% 
  mutate(total_mean = mean(total_sizefraction)) %>% 
  ungroup() %>% 
  distinct(sampletype_f, sizefraction_f, labid, total_mean) %>%
  group_by(sampletype_f, sizefraction_f) %>% 
  mutate(`Mean`= mean(total_mean)) %>%  
  mutate(`Median` = median(total_mean)) %>%
  mutate(`SD` = sd(total_mean)) %>%
  mutate(`Min` = min(total_mean)) %>% 
  mutate(`Max` = max(total_mean)) %>%
  ungroup() %>% 
  select(sampletype_f, sizefraction_f, `Mean`, `Median`, `SD`, `Min`, `Max`) %>%
  rename("Matrix" = sampletype_f) %>%
  rename("Size Fraction" = sizefraction_f) %>%
  distinct() 

kable(f, digits = 2, caption = "Raw Particle Count: Size Fraction. Global summary statistics. Not blank corrected.")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

#Color Count
g <- Results_Clean_Microscopy %>%  
  filter(!grepl("4", sampleid)) %>%
  filter(!grepl("8", sampleid)) %>% #Dropping blanks
  group_by(sampletype_f, color_f, labid, set, total_color) %>% 
  summarise() %>% 
  mutate(total_mean = mean(total_color)) %>% 
  ungroup() %>% 
  distinct(sampletype_f, color_f, labid, total_mean) %>%
  group_by(sampletype_f, color_f) %>% 
  mutate(`Mean`= mean(total_mean)) %>%  
  mutate(`Median` = median(total_mean)) %>%
  mutate(`SD` = sd(total_mean)) %>%
  mutate(`Min` = min(total_mean)) %>% 
  mutate(`Max` = max(total_mean)) %>%
  ungroup() %>% 
  select(sampletype_f, color_f, `Mean`, `Median`, `SD`, `Min`, `Max`) %>%
  rename("Matrix" = sampletype_f) %>%
  rename("Color" = color_f) %>%
  distinct() 

kable(g, digits = 2, caption = "Raw Particle Count: Color. Global summary statistics. Not blank corrected.")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

#Morphology Count
h <- Results_Clean_Microscopy %>%  
  filter(!grepl("4", sampleid)) %>%
  filter(!grepl("8", sampleid)) %>% #Dropping blanks
  group_by(sampletype_f, morphology_f, labid, set, total_morphology) %>% 
  summarise() %>% 
  mutate(total_mean = mean(total_morphology)) %>% 
  ungroup() %>% 
  distinct(sampletype_f, morphology_f, labid, total_mean) %>%
  group_by(sampletype_f, morphology_f) %>% 
  mutate(`Mean`= mean(total_mean)) %>%  
  mutate(`Median` = median(total_mean)) %>%
  mutate(`SD` = sd(total_mean)) %>%
  mutate(`Min` = min(total_mean)) %>% 
  mutate(`Max` = max(total_mean)) %>%
  ungroup() %>% 
  select(sampletype_f, morphology_f, `Mean`, `Median`, `SD`, `Min`, `Max`) %>%
  rename("Matrix" = sampletype_f) %>%
  rename("Morphology" = morphology_f) %>%
  distinct() 

kable(h, digits = 2, caption = "Raw Particle Count: Morphology. Global summary statistics. Not blank corrected.")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

```


## Recovery

The following plots display the recovery (reported count/spiked count (%)) obtained from microscopy. **Reported particle characteristics must match expected particle characteristics of the spiking materials to count towards recovery estimations.** For example, the colors of polymer spiking materials are defined as white, orange, blue, clear and green. Colors outside of these definitions are not included in this section (see Counts). 

The dotted line is at 100% and the grey area represents the mean RSD for each given category. The data are **not** blank corrected.

Recovery in this section is estimated based on a single particle characteristic. For example, if the sample was spiked with 10 green particles and 8 green particles were recovered, recovery is estimated as 80% regardless of whether or not the shape or size fraction also matches the spiking material. More restrictive recovery analysis are reported later on in this document.

### Total

```{r echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, fig.width = 12, fig.height = 10}

#Total - By Lab
a <- Join_Microscopy_Total %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%  
  #Creating plot
  ggplot(aes(x=labid, y= total_pct_recovery)) +
  theme_test() +
  geom_rect(ymin = 95.7, ymax = 104.3, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Total Particle Recovery", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(a)

#Total - By Lab
a_a <- Join_Microscopy_Total %>% 
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  ungroup() %>% 
  group_by(labid) %>% 
  mutate(All_Matrices = if_else(n_distinct(sampletype_f) == 4, "Y", "N")) %>% 
  filter(All_Matrices == "Y") %>% 
  ungroup() %>%  
  #Creating plot
  ggplot(aes(x=labid, y= total_pct_recovery)) +
  theme_test() +
  geom_rect(ymin = 95.7, ymax = 104.3, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = sampletype_f, fill = sampletype_f, group = sampletype_f), alpha = 0.7, width = .05, size = 3, position = position_dodge(width = 0.75)) +
  geom_boxplot(alpha = 0.5,aes(color = sampletype_f, fill =sampletype_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4"))+
  scale_color_manual(values = c("royalblue", "forestgreen", "darksalmon", "goldenrod4"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Matrix", color = "Matrix", title = "Microscopy", subtitle = "Total Particle Recovery", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(a_a)

b <- Join_Microscopy_Total %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  group_by(sampletype_f, labid, sampleid, total_pct_recovery) %>%
  summarise() %>% 
  rename("Matrix" = sampletype_f) %>% 
  rename("Lab ID" = labid) %>%
  rename("Sample ID" = sampleid) %>%
  rename("Recovery" = total_pct_recovery) %>%
  distinct() 

kable(b, digits = 2, caption = "Microscopy: Total Particle Recovery. Summary statistics by sample (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

#Total - Summary
c<- Join_Microscopy_Total %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, total_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid) %>% 
  mutate(total_pct_recovery_mean = mean(total_pct_recovery)) %>% 
  ungroup() %>% 
  group_by(sampletype_f, labid, microscopy_f, total_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= total_pct_recovery_mean)) +
  theme_test() +
  geom_rect(ymin = 95.7, ymax = 104.3, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Total Particle Recovery", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(c)

d <- Join_Microscopy_Total %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  group_by(sampletype_f, labid, set) %>% 
  mutate(`Mean (%)`= mean(total_pct_recovery)) %>%  
  mutate(`Median (%)` = median(total_pct_recovery)) %>%
  mutate(`SD (%)` = sd(total_pct_recovery)) %>%
  mutate(`SE (%)` = std.error(total_pct_recovery)) %>%
  mutate(`Min (%)` = min(total_pct_recovery)) %>% 
  mutate(`Max (%)` = max(total_pct_recovery)) %>%
  ungroup() %>%
  group_by(sampletype_f) %>% 
  mutate(mean_se = mean(`SE (%)`, na.rm = TRUE)) %>% 
  select(sampletype_f, labid, `Mean (%)`, `Median (%)`, `SD (%)`, `SE (%)`, `Min (%)`, `Max (%)`, mean_se) %>%
  rename("Matrix" = sampletype_f) %>%  
  rename("Lab ID" = labid) %>%
  distinct() 

kable(d, digits = 2, caption = "Microscopy: Total Particle Recovery. Summary statistics by laboratory (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

e <- Join_Microscopy_Total %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  group_by(sampletype_f, labid, set) %>% 
  mutate(mean_lab = mean(total_pct_recovery)) %>% 
  ungroup() %>%
  distinct(sampletype_f, labid, mean_lab) %>%
  group_by(sampletype_f) %>% 
  mutate(`Mean (%)`= mean(mean_lab)) %>%    
  mutate(`Median (%)` = median(mean_lab)) %>%
  mutate(`SD (%)` = sd(mean_lab)) %>%
  mutate(`SE (%)` = std.error(mean_lab)) %>%
  mutate(`Min (%)` = min(mean_lab)) %>% 
  mutate(`Max (%)` = max(mean_lab)) %>%
  ungroup() %>% 
  rename("Matrix" = sampletype_f) %>% 
  select(`Matrix`, `Mean (%)`, `Median (%)`, `SD (%)`, `SE (%)`, `Min (%)`, `Max (%)`) %>%
  distinct() 

kable(e, digits = 2, caption = "Microscopy: Total Particle Recovery. Summary statistics across all laboratories. Not blank corrected.")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")
```

```{r Manuscript Fig 1C, echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, include = FALSE}

#Total - Summary
Fig1C<- Join_Microscopy_Total %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  mutate(sampletype_f = case_when(
    sampletype_f == "Clean Water" ~ "Drinking Water",
    sampletype_f == "Dirty Water" ~ "Surface Water",
    sampletype_f == "Fish Tissue" ~ "Fish Tissue",
    sampletype_f == "Sediment" ~ "Sediment")) %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, total_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid) %>% 
  mutate(total_pct_recovery_mean = mean(total_pct_recovery)) %>% 
  ungroup() %>% 
  group_by(sampletype_f, labid, microscopy_f, total_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=sampletype_f, y= total_pct_recovery_mean)) +
  theme_test() +
  geom_point(aes(color = sampletype_f, fill = sampletype_f, alpha = 0.7), size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = sampletype_f, fill = sampletype_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  theme(text = element_text(size = 20),
        plot.title = element_text(hjust = 0.5, vjust = -6),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("royalblue", "darksalmon", "goldenrod4", "forestgreen")) +   
  scale_fill_manual(values = c("royalblue", "darksalmon", "goldenrod4", "forestgreen")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", title = "> 212 µm")

plot(Fig1C)

saveRDS(Fig1C, file = "./Manuscript_Figures/Fig1C")

```

```{r Manuscript Fig S10, echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, include = FALSE}

#Total - By Lab
FigS10 <- Join_Microscopy_Total %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  ungroup() %>% 
  mutate(sampletype_f = case_when(
    sampletype_f == "Clean Water" ~ "Drinking Water",
    sampletype_f == "Dirty Water" ~ "Surface Water",
    sampletype_f == "Fish Tissue" ~ "Fish Tissue",
    sampletype_f == "Sediment" ~ "Sediment")) %>%
  #Creating plot
  ggplot(aes(x=labid, y= total_pct_recovery)) +
  theme_test() +
  geom_point(aes(color = sampletype_f, fill = sampletype_f, alpha = 0.7), size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 20),
        plot.title = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darksalmon", "goldenrod4", "forestgreen"))+
  scale_color_manual(values = c("royalblue", "darksalmon", "goldenrod4", "forestgreen"))+
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", x = "Lab", title = "> 212 µm")

plot(FigS10)

saveRDS(FigS10, file = "./Manuscript_Figures/FigS10")

```

### Size Fraction

```{r echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, fig.width = 12, fig.height = 10}

#All Size Fractions
a <- Join_Microscopy_SizeFraction %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Force size fractions to be in the correct order
  mutate(sizefraction_f = factor(sizefraction_f, levels = c("212-500 µm",">500 µm"))) %>% 
  #Data manipulation
  group_by(sampletype_f, labid, sampleid, microscopy_f, sizefraction_f, sizefrac_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, sizefraction_f) %>% 
  mutate(sizefrac_pct_recovery_mean = mean(sizefrac_pct_recovery)) %>% 
  ungroup() %>% 
  group_by(sampletype_f, labid, microscopy_f, sizefraction_f, sizefrac_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=sampletype_f, y= sizefrac_pct_recovery_mean)) +
  theme_test() +
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sizefraction_f, ncol = 2, scales = "free")+
  theme(text = element_text(size = 13),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, All Size Fractions", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(a)

#212-500 µm - By Lab
f <- Join_Microscopy_SizeFraction %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  filter(sizefraction_f == "212-500 µm") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= sizefrac_pct_recovery)) +
  theme_test() +
  geom_rect(ymin = 91.05, ymax = 108.95, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, 212-500 µm", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(f)

#212-500 µm - Summary
g <- Join_Microscopy_SizeFraction %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, sizefraction_f, sizefrac_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, sizefraction_f) %>% 
  mutate(sizefrac_pct_recovery_mean = mean(sizefrac_pct_recovery)) %>% 
  ungroup() %>% 
  filter(sizefraction_f == "212-500 µm") %>%
  group_by(sampletype_f, labid, microscopy_f, sizefrac_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= sizefrac_pct_recovery_mean)) +
  theme_test() +
  geom_rect(ymin = 91.05, ymax = 108.95, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, 212-500 µm", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(g)

#>500 µm - By Lab
h <- Join_Microscopy_SizeFraction %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  filter(sizefraction_f == ">500 µm") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= sizefrac_pct_recovery)) +
  theme_test() +
  #No RSD data for the >500 µm size fraction
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, >500 µm", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(h)

#>500 µm - Summary
i <- Join_Microscopy_SizeFraction %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, sizefraction_f, sizefrac_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, sizefraction_f) %>% 
  mutate(sizefrac_pct_recovery_mean = mean(sizefrac_pct_recovery)) %>% 
  ungroup() %>% 
  filter(sizefraction_f == ">500 µm") %>%
  group_by(sampletype_f, labid, microscopy_f, sizefrac_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= sizefrac_pct_recovery_mean)) +
  theme_test() +
  #No RSD data for the >500 µm size fraction
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, >500 µm", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(i)

#Size Fraction - By Sample
j <- Join_Microscopy_SizeFraction %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  group_by(sampletype_f, labid, sampleid, sizefraction_f, sizefrac_pct_recovery) %>% 
  summarise() %>% 
  rename("Matrix" = sampletype_f) %>%
  rename("Lab ID" = labid) %>%
  rename("Sample ID" = sampleid) %>%
  rename("Size Fraction" = sizefraction_f) %>%
  rename("Recovery" = sizefrac_pct_recovery) %>%
  distinct() %>% 
  arrange(`Matrix`, `Lab ID`, `Sample ID`, `Size Fraction`)

kable(j, digits = 2, caption = "Microscopy: Total Particle Recovery by Size Fraction. Summary statistics by sample (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

#Size Fraction - By Lab
k <- Join_Microscopy_SizeFraction %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>% 
  group_by(sampletype_f, labid, set, sizefraction_f) %>% 
  mutate(`Mean (%)`= mean(sizefrac_pct_recovery)) %>%  
  mutate(`Median (%)` = median(sizefrac_pct_recovery)) %>%
  mutate(`SD (%)` = sd(sizefrac_pct_recovery)) %>%
  mutate(`SE (%)` = std.error(sizefrac_pct_recovery)) %>%
  mutate(`Min (%)` = min(sizefrac_pct_recovery)) %>% 
  mutate(`Max (%)` = max(sizefrac_pct_recovery)) %>%
  ungroup() %>% 
  select(sampletype_f, labid, sizefraction_f, `Mean (%)`, `Median (%)`, `SD (%)`, `SE (%)`, `Min (%)`, `Max (%)`) %>%
  rename("Matrix" = sampletype_f) %>%
  rename("Lab ID" = labid) %>%
  rename("Size Fraction" = sizefraction_f) %>%
  distinct() %>% 
  arrange(`Lab ID`, `Size Fraction`)

kable(k, digits = 2, caption = "Microscopy: Total Particle Recovery by Size Fraction. Summary statistics by laboratory (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

#Size Fraction - Summary
l <- k %>% 
  group_by(`Matrix`, `Size Fraction`) %>% 
  mutate(`Global Mean (%)`= mean(`Mean (%)`)) %>%   
  mutate(`Median (%)` = median(`Mean (%)`)) %>%
  mutate(`SD (%)` = sd(`Mean (%)`)) %>%
  mutate(`SE (%)` = std.error(`Mean (%)`)) %>%
  mutate(`Min (%)` = min(`Mean (%)`)) %>% 
  mutate(`Max (%)` = max(`Mean (%)`)) %>%
  ungroup() %>% 
  select(`Matrix`, `Size Fraction`, `Global Mean (%)`, `Median (%)`, `SD (%)`, `SE (%)`, `Min (%)`, `Max (%)`) %>%
  distinct() %>% 
  arrange(`Matrix`,`Size Fraction`)

kable(l, digits = 2, caption = "Microscopy: Total Particle Recovery by Size Fraction. Summary statistics across all laboratories. Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

```

```{r Manuscript Fig S11, echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, include = FALSE}

#All Size Fractions
FigS11 <- Join_Microscopy_SizeFraction %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  ungroup() %>%
  mutate(sampletype_f = case_when(
    sampletype_f == "Clean Water" ~ "Drinking Water",
    sampletype_f == "Dirty Water" ~ "Surface Water",
    sampletype_f == "Fish Tissue" ~ "Fish Tissue",
    sampletype_f == "Sediment" ~ "Sediment")) %>%
  #Force size fractions to be in the correct order
  mutate(sizefraction_f = factor(sizefraction_f, levels = c("212-500 µm",">500 µm"))) %>% 
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, sizefraction_f, sizefrac_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, sizefraction_f) %>% 
  mutate(sizefrac_pct_recovery_mean = mean(sizefrac_pct_recovery)) %>% 
  ungroup() %>% 
  group_by(sampletype_f, labid, microscopy_f, sizefraction_f, sizefrac_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=sampletype_f, y= sizefrac_pct_recovery_mean)) +
  theme_test() +
  geom_point(aes(color = sampletype_f, fill = sampletype_f), alpha = 0.7, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = sampletype_f, fill = sampletype_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sizefraction_f, nrow = 1, scales = "free")+
  theme(text = element_text(size = 20),
        legend.position = "right",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("royalblue", "darksalmon", "goldenrod4", "forestgreen")) +   
  scale_fill_manual(values = c("royalblue", "darksalmon", "goldenrod4", "forestgreen")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Matrix", color = "Matrix")

plot(FigS11)

saveRDS(FigS11, file = "./Manuscript_Figures/FigS11")

```

### Color

```{r echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, fig.width = 12, fig.height = 10}

#All Colors
a <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, color_f, color_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, color_f) %>% 
  mutate(color_pct_recovery_mean = mean(color_pct_recovery)) %>% 
  ungroup() %>% 
  group_by(sampletype_f, labid, microscopy_f, color_f, color_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=sampletype_f, y= color_pct_recovery_mean)) +
  theme_test() +
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.7,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~color_f, ncol = 3, scales = "free")+
  theme(text = element_text(size = 13),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", subtitle = "Particle Recovery, All Colors", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(a)

#Blue - By Lab
b <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  filter(color_f == "Blue") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= color_pct_recovery)) +
  theme_test() +
  #No RSD data for blue particles
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, Blue", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(b)

#Blue - Summary
c <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, color_f, color_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, color_f) %>% 
  mutate(color_pct_recovery_mean = mean(color_pct_recovery)) %>% 
  ungroup() %>% 
  filter(color_f == "Blue") %>%
  group_by(sampletype_f, labid, microscopy_f, color_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= color_pct_recovery_mean)) +
  theme_test() +
  #No RSD data for blue particles
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, Blue", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(c)

#White - By Lab
d <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  filter(color_f == "White") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= color_pct_recovery)) +
  theme_test() +
  geom_rect(ymin = 93.5, ymax = 106.5, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, White", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(d)

#White - Summary
e <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, color_f, color_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, color_f) %>% 
  mutate(color_pct_recovery_mean = mean(color_pct_recovery)) %>% 
  ungroup() %>% 
  filter(color_f == "White") %>%
  group_by(sampletype_f, labid, microscopy_f, color_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= color_pct_recovery_mean)) +
  theme_test() +
  geom_rect(ymin = 93.5, ymax = 106.5, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, White", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(e)

#Green - By Lab
f <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  filter(color_f == "Green") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= color_pct_recovery)) +
  theme_test() +
  #No RSD data for green particles
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, Green", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(f)

#Green - Summary
g <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, color_f, color_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, color_f) %>% 
  mutate(color_pct_recovery_mean = mean(color_pct_recovery)) %>% 
  ungroup() %>% 
  filter(color_f == "Green") %>%
  group_by(sampletype_f, labid, microscopy_f, color_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= color_pct_recovery_mean)) +
  theme_test() +
  #No RSD data for green particles
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, Green", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(g)

#Orange - By Lab
h <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  filter(color_f == "Orange") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= color_pct_recovery)) +
  theme_test() +
  #No RSD data for orange particles
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, Orange", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(h)

#Orange - Summary
i <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, color_f, color_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, color_f) %>% 
  mutate(color_pct_recovery_mean = mean(color_pct_recovery)) %>% 
  ungroup() %>% 
  filter(color_f == "Orange") %>%
  group_by(sampletype_f, labid, microscopy_f, color_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= color_pct_recovery_mean)) +
  theme_test() +
  #No RSD data for orange particles
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, Orange", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(i)

#Clear - By Lab
j <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  filter(color_f == "Clear") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= color_pct_recovery)) +
  theme_test() +
  #No RSD data for Clear particles
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, Clear", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(j)

#Clear - Summary
k <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, color_f, color_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, color_f) %>% 
  mutate(color_pct_recovery_mean = mean(color_pct_recovery)) %>% 
  ungroup() %>% 
  filter(color_f == "Clear") %>%
  group_by(sampletype_f, labid, microscopy_f, color_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= color_pct_recovery_mean)) +
  theme_test() +
  #No RSD data for Clear particles
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, Clear", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(k)

#Color - By Sample
l <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  group_by(sampletype_f, labid, sampleid, color_f, color_pct_recovery) %>% 
  summarise() %>% 
  rename("Matrix" = sampletype_f) %>%
  rename("Lab ID" = labid) %>%
  rename("Sample ID" = sampleid) %>%
  rename("Color" = color_f) %>%
  rename("Recovery" = color_pct_recovery) %>%
  distinct() 

kable(l, digits = 2, caption = "Microscopy: Total Particle Recovery by Size Fraction. Summary statistics by sample (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

#Color - By Lab
m <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  group_by(sampletype_f, labid, set, color_f) %>% 
  mutate(`Mean (%)`= mean(color_pct_recovery)) %>%  
  mutate(`Median (%)` = median(color_pct_recovery)) %>%
  mutate(`SD (%)` = sd(color_pct_recovery)) %>%
  mutate(`SE (%)` = std.error(color_pct_recovery)) %>%
  mutate(`Min (%)` = min(color_pct_recovery)) %>% 
  mutate(`Max (%)` = max(color_pct_recovery)) %>%
  ungroup() %>% 
  select(sampletype_f, labid, color_f, `Mean (%)`, `Median (%)`, `SD (%)`, `SE (%)`, `Min (%)`, `Max (%)`) %>%
  rename("Matrix" = sampletype_f) %>%
  rename("Lab ID" = labid) %>%
  rename("Color" = color_f) %>%
  distinct() 

kable(m, digits = 2, caption = "Microscopy: Total Particle Recovery by Color. Summary statistics by laboratory (n = 1-3 samples). Not blank corrected.")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

#Color - Summary
n <- m %>%
  group_by(`Matrix`, `Color`) %>% 
  mutate(`Global Mean (%)`= mean(`Mean (%)`)) %>%   
  mutate(`Median (%)` = median(`Mean (%)`)) %>%
  mutate(`SD (%)` = sd(`Mean (%)`)) %>%
  mutate(`SE (%)` = std.error(`Mean (%)`)) %>%
  mutate(`Min (%)` = min(`Mean (%)`)) %>% 
  mutate(`Max (%)` = max(`Mean (%)`)) %>%
  ungroup() %>% 
  select(`Matrix`, `Color`, `Global Mean (%)`, `Median (%)`, `SD (%)`, `SE (%)`, `Min (%)`, `Max (%)`) %>%
  distinct() 

kable(n, digits = 2, caption = "Microscopy: Total Particle Recovery by Color. Summary statistics across all laboratories. Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")
  
```

```{r Manuscript Fig S12, echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, include = FALSE}

#All Colors
FigS12 <- Join_Microscopy_Color %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  mutate(sampletype_f = case_when(
    sampletype_f == "Clean Water" ~ "Drinking Water",
    sampletype_f == "Dirty Water" ~ "Surface Water",
    sampletype_f == "Fish Tissue" ~ "Fish Tissue",
    sampletype_f == "Sediment" ~ "Sediment")) %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, color_f, color_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, color_f) %>% 
  mutate(color_pct_recovery_mean = mean(color_pct_recovery)) %>% 
  ungroup() %>% 
  group_by(sampletype_f, labid, microscopy_f, color_f, color_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=sampletype_f, y= color_pct_recovery_mean)) +
  theme_test() +
  geom_point(aes(color = sampletype_f, fill = sampletype_f), alpha = 0.7, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = sampletype_f, fill = sampletype_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~color_f, nrow = 1, scales = "free")+
  theme(text = element_text(size = 20),
        legend.position = "right",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("royalblue", "darksalmon", "goldenrod4", "forestgreen")) +   
  scale_fill_manual(values = c("royalblue", "darksalmon", "goldenrod4", "forestgreen")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Matrix", color = "Matrix") 

plot(FigS12)

saveRDS(FigS12, file = "./Manuscript_Figures/FigS12")

```

### Morphology

```{r echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, fig.width = 12, fig.height = 10}

#All Morphologies
a <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, morphology_f, morphology_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, morphology_f) %>% 
  mutate(morphology_pct_recovery_mean = mean(morphology_pct_recovery)) %>% 
  ungroup() %>% 
  group_by(sampletype_f, labid, microscopy_f, morphology_f, morphology_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=sampletype_f, y= morphology_pct_recovery_mean)) +
  theme_test() +
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.7,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~morphology_f, ncol = 3, scales = "free")+
  theme(text = element_text(size = 13),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, All Morphologies", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(a)


#Fibers - By Lab
b <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  filter(morphology_f == "Fiber") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= morphology_pct_recovery)) +
  theme_test() +
  #No RSD data for fibers
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, Fibers", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(b)

#Fiber - Summary
c <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, morphology_f, morphology_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, morphology_f) %>% 
  mutate(morphology_pct_recovery_mean = mean(morphology_pct_recovery)) %>% 
  ungroup() %>% 
  filter(morphology_f == "Fiber") %>%
  group_by(sampletype_f, labid, microscopy_f,morphology_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= morphology_pct_recovery_mean)) +
  theme_test() +
  #No RSD data for fibers
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, Fibers", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(c)

#Spheres - By Lab
d <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  filter(morphology_f == "Sphere") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= morphology_pct_recovery)) +
  theme_test() +
  geom_rect(ymin = 98.2, ymax = 101.8, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, Spheres", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(d)

#Sphere - Summary
e <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, morphology_f, morphology_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, morphology_f) %>% 
  mutate(morphology_pct_recovery_mean = mean(morphology_pct_recovery)) %>% 
  ungroup() %>% 
  filter(morphology_f == "Sphere") %>%
  group_by(sampletype_f, labid, microscopy_f,morphology_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= morphology_pct_recovery_mean)) +
  theme_test() +
  geom_rect(ymin = 98.2, ymax = 101.8, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, Sphere", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(e)

#Fragments - By Lab
f <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  filter(morphology_f == "Fragment") %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%
  #Creating plot
  ggplot(aes(x=labid, y= morphology_pct_recovery)) +
  theme_test() +
  geom_rect(ymin = 93.6, ymax = 106.4, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", subtitle = "Particle Recovery, Fragments", caption = "Points represent individual samples spiked with microplastic particles.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(f)

#Fragments - Summary
g <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, morphology_f, morphology_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, morphology_f) %>% 
  mutate(morphology_pct_recovery_mean = mean(morphology_pct_recovery)) %>% 
  ungroup() %>% 
  filter(morphology_f == "Fragment") %>%
  group_by(sampletype_f, labid, microscopy_f,morphology_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=microscopy_f, y= morphology_pct_recovery_mean)) +
  theme_test() +
  geom_rect(ymin = 93.6, ymax = 106.4, xmin = 0, xmax = Inf, color = "gray87", fill = "gray87", alpha = 0.5)+
  geom_point(aes(color = microscopy_f, fill = microscopy_f, alpha = 0.7), width = .05, size = 3) +
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f, fill = microscopy_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~sampletype_f, ncol = 2)+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("darkmagenta")) +   
  scale_fill_manual(values = c("darkmagenta")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Lab", title = "Microscopy", subtitle = "Particle Recovery, Fragments", caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line and shaded area represent 100% recovery of polymer materials and RSD, respectively.\nNot blank corrected.")

plot(g)

#Morphology - By Sample
h <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  group_by(sampletype_f, labid, sampleid, morphology_f, morphology_pct_recovery) %>% 
  summarise() %>%
  rename("Matrix" = sampletype_f) %>% 
  rename("Lab ID" = labid) %>%
  rename("Sample ID" = sampleid) %>%
  rename("Morphology" = morphology_f) %>%
  rename("Recovery" = morphology_pct_recovery) %>%
  distinct() 

kable(h, digits = 2, caption = "Microscopy: Total Particle Recovery by Size Fraction. Summary statistics by sample (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")

#Morphology - By Lab
i <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  group_by(sampletype_f, labid, set, morphology_f) %>% 
  mutate(`Mean (%)`= mean(morphology_pct_recovery)) %>%  
  mutate(`Median (%)` = median(morphology_pct_recovery)) %>%
  mutate(`SD (%)` = sd(morphology_pct_recovery)) %>%
  mutate(`SE (%)` = std.error(morphology_pct_recovery)) %>%
  mutate(`Min (%)` = min(morphology_pct_recovery)) %>% 
  mutate(`Max (%)` = max(morphology_pct_recovery)) %>%
  ungroup() %>% 
  select(sampletype_f, labid, morphology_f, `Mean (%)`, `Median (%)`, `SD (%)`, `SE (%)`, `Min (%)`, `Max (%)`) %>%
  rename("Matrix" = sampletype_f) %>% 
  rename("Lab ID" = labid) %>%
  rename("Morphology" = morphology_f) %>%
  distinct() 

kable(i, digits = 2, caption = "Microscopy: Total Particle Recovery by Morphology. Summary statistics by laboratory (n = 1-3 samples). Not blank corrected.") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px") 

#Morphology - Summary
j <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  group_by(sampletype_f, morphology_f) %>% 
  mutate(`Mean (%)`= mean(morphology_pct_recovery)) %>%  
  mutate(`Median (%)` = median(morphology_pct_recovery)) %>%
  mutate(`SD (%)` = sd(morphology_pct_recovery)) %>%
  mutate(`SE (%)` = std.error(morphology_pct_recovery)) %>%
  mutate(`Min (%)` = min(morphology_pct_recovery)) %>% 
  mutate(`Max (%)` = max(morphology_pct_recovery)) %>%
  ungroup() %>% 
  select(sampletype_f, morphology_f, `Mean (%)`, `Median (%)`, `SD (%)`, `SE (%)`, `Min (%)`, `Max (%)`) %>%
  rename("Matrix" = sampletype_f) %>%
  rename("Morphology" = morphology_f) %>%
  distinct() 

kable(j, digits = 2, caption = "Microscopy: Total Particle Recovery by Morphology. Summary statistics across all laboratories. Not blank corrected.")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>% 
  scroll_box(width = "700px", height = "200px")
```

```{r Manuscript Fig S13, echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, include = FALSE}

FigS13 <- Join_Microscopy_Morphology %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f != "N") %>%
  mutate(sampletype_f = case_when(
    sampletype_f == "Clean Water" ~ "Drinking Water",
    sampletype_f == "Dirty Water" ~ "Surface Water",
    sampletype_f == "Fish Tissue" ~ "Fish Tissue",
    sampletype_f == "Sediment" ~ "Sediment")) %>%
  #Data manipulation
  group_by(sampletype_f, labid, set, sampleid, microscopy_f, morphology_f, morphology_pct_recovery) %>% 
  summarise() %>% 
  group_by(sampletype_f, labid, morphology_f) %>% 
  mutate(morphology_pct_recovery_mean = mean(morphology_pct_recovery)) %>% 
  ungroup() %>% 
  group_by(sampletype_f, labid, microscopy_f, morphology_f, morphology_pct_recovery_mean) %>% 
  summarise() %>% 
  #Creating plot
  ggplot(aes(x=sampletype_f, y= morphology_pct_recovery_mean)) +
  theme_test() +
  geom_point(aes(color = sampletype_f, fill = sampletype_f), alpha = 0.7, size = 3) +
  geom_boxplot(alpha = 0.7,aes(color = sampletype_f, fill = sampletype_f), outlier.shape = NA) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  facet_wrap(~morphology_f, ncol = 3, scales = "free")+
  theme(text = element_text(size = 20),
        legend.position = "right",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  scale_color_manual(values = c("royalblue", "darksalmon", "goldenrod4", "forestgreen")) +   
  scale_fill_manual(values = c("royalblue", "darksalmon", "goldenrod4", "forestgreen")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Matrix", color = "Matrix") 

plot(FigS13)

saveRDS(FigS13, file = "./Manuscript_Figures/FigS13")

```

### Color and Morphology within Size Fraction

The following plots display the recovery (reported count/spiked count (%)) obtained from microscopy. **Reported particle characteristics must match expected particle characteristics of the spiking materials to count towards recovery estimations.** For example, the colors of polymer spiking materials are defined as white, orange, blue, clear and green. Colors outside of these definitions are currently excluded from analysis. 

The dotted line is at 100% and the grey area represents the mean RSD for each given category. The data are **not** blank corrected.

Recovery in this section is estimated based on color and morphology within a size fraction. For example, if the sample was spiked with 10 green spheres >500 µm and 8 green spheres were recovered in the >500 µm size fraction, recovery is estimated as 80%. However, if 4 of these spheres were reported as blue rather than green, recovery would fall to 60%. 

```{r echo = FALSE, error = FALSE, warning = FALSE, message = FALSE, results = 'asis'}

#Make data frame of particle characteristics
df <- Join_Microscopy_Combo %>%
  filter(!(labid == "R"&sampletype_f == "Clean Water")) %>%
  filter(plastic_f == "Y") %>% 
  group_by(sampletype_f, morphology_f, color_f, sizefraction_f) %>% 
  summarise()

for (i in 1:nrow(df))

{ #Open loop  
  
#By Lab
a <- Join_Microscopy_Combo %>%
  group_by(sampletype_f, labid, sampleid, microscopy_f,sizefraction_f, color_f, morphology_f, color_morphology_sizefraction_pct_recovery) %>% 
  filter(sizefraction_f == df$sizefraction_f[[i]]) %>% 
  filter(color_f == df$color_f[[i]]) %>% 
  filter(morphology_f == df$morphology_f[[i]]) %>%
  ungroup() %>% 
  mutate(labid = fct_reorder(labid, expertise)) %>%  
  #Creating plot
  ggplot(aes(x=labid, y= color_morphology_sizefraction_pct_recovery)) +
  theme_test() +
  geom_point(aes(color = as.factor(expertise), fill = as.factor(expertise)), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. - 0.25, yend=..y..))+
  stat_summary(fun="mean", geom="segment", mapping=aes(xend=..x.. + 0.25, yend=..y..))+
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "right",
        axis.text.x = element_text(angle = 45, vjust = 1, hjust=1),
        axis.ticks.x = element_blank())+
  facet_wrap(~sampletype_f, scales = "free")+
  scale_fill_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  scale_color_manual(values = c("royalblue", "darkcyan", "darkblue", "orchid"))+
  expand_limits(y = 0) +
  labs(x = "Labs", y = "Recovery (%)", fill = "Experience", color = "Experience", title = "Microscopy", 
       subtitle = paste(df$sizefraction_f[[i]], df$color_f[[i]], df$morphology_f[[i]]), 
       caption = "Points represent individual samples spiked with microplastic particles.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(a)

#Summary
b <- Join_Microscopy_Combo %>%
  filter(sizefraction_f == df$sizefraction_f[[i]]) %>% 
  filter(color_f == df$color_f[[i]]) %>% 
  filter(morphology_f == df$morphology_f[[i]]) %>%
  group_by(sampletype_f, labid) %>%
  mutate(color_morphology_sizefraction_pct_recovery_mean = mean(color_morphology_sizefraction_pct_recovery)) %>%
  ungroup() %>%
  group_by(sampletype_f, labid, microscopy_f,color_morphology_sizefraction_pct_recovery_mean) %>%
  summarise() %>%
  #Creating plot
  ggplot(aes(x=microscopy_f,y= color_morphology_sizefraction_pct_recovery_mean)) +
  theme_test() +
  geom_point(aes(color = microscopy_f, fill = microscopy_f), alpha = 0.7, width = .05, size = 3) +
  geom_hline(yintercept = 100, linetype = "dotted")+
  geom_boxplot(alpha = 0.5,aes(color = microscopy_f,fill = microscopy_f), outlier.shape = NA) +
  theme(text = element_text(size = 12),
        plot.title = element_text(hjust = 0.5),
        plot.subtitle = element_text(hjust = 0.5),
        legend.position = "none",
        axis.text.x = element_blank(),
        axis.title.x = element_blank(),
        axis.ticks.x = element_blank())+
  facet_wrap(~sampletype_f, scales = "free")+
  scale_color_manual(values = c("darkmagenta", "orchid")) +
  scale_fill_manual(values = c("darkmagenta", "orchid")) +
  expand_limits(y = 0) +
  labs(y = "Recovery (%)", fill = "Method", color = "Method", title = "Microscopy", 
       subtitle = paste(df$sizefraction_f[[i]], df$color_f[[i]], df$morphology_f[[i]]),
       caption = "Points represent the sample mean (n = 1-3) of each laboratory.\nDotted line represents 100% recovery of polymer materials.\nNot blank corrected.")

plot(b)
 
#By Sample
c <- Join_Microscopy_Combo %>%
  group_by(sampletype_f, labid, sizefraction_f, color_f, morphology_f) %>%
  filter(sizefraction_f == df$sizefraction_f[[i]]) %>% 
  filter(color_f == df$color_f[[i]]) %>% 
  filter(morphology_f == df$morphology_f[[i]]) %>%
  ungroup() %>%
  select(sampletype_f, labid, sampleid, color_morphology_sizefraction_pct_recovery) %>%
  rename("Matrix" = sampletype_f) %>% 
  rename("Lab ID" = labid) %>%
  rename("Sample ID" = sampleid) %>%
  rename("Recovery" = color_morphology_sizefraction_pct_recovery) %>%
  distinct()

print(kable(c, digits = 2, caption = paste(df$sizefraction_f[[i]], df$color_f[[i]], df$morphology_f[[i]])) %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px"))
  cat('\n')

#By Lab
d <- Join_Microscopy_Combo %>%
  group_by(sampletype_f, labid, sizefraction_f, color_f, morphology_f) %>%
  filter(sizefraction_f == df$sizefraction_f[[i]]) %>% 
  filter(color_f == df$color_f[[i]]) %>% 
  filter(morphology_f == df$morphology_f[[i]]) %>%
  mutate(Mean = mean(color_morphology_sizefraction_pct_recovery)) %>%
  mutate(Median = median(color_morphology_sizefraction_pct_recovery)) %>%
  mutate(SD = sd(color_morphology_sizefraction_pct_recovery)) %>%
  mutate(SE = std.error(color_morphology_sizefraction_pct_recovery)) %>%
  mutate(Min = min(color_morphology_sizefraction_pct_recovery)) %>%
  mutate(Max = max(color_morphology_sizefraction_pct_recovery)) %>%
  ungroup() %>%
  select(sampletype_f, labid, Mean, Median, SD, SE, Min, Max) %>%
  rename("Matrix" = sampletype_f) %>% 
  rename("Lab ID" = labid) %>%
  rename("Mean (%)" = Mean) %>%
  rename("Median (%)" = Median) %>%
  rename("SD (%)" = SD) %>%
  rename("SE (%)" = SE) %>%
  rename("Min (%)" = Min) %>%
  rename("Max (%)" = Max) %>%
  distinct()

print(kable(d, digits = 2, caption = paste(df$sizefraction_f[[i]], df$color_f[[i]], df$morphology_f[[i]])) %>% 
  kable_styling(bootstrap_options = c("striped", "hover", "condensed")) %>%
  scroll_box(width = "700px", height = "200px"))  
  cat('\n')

 } #Close loop

```

End of Analysis.