June 2023 Citizen Science Research

During this survey period our teams in Falmouth and Plymouth conducted mini-biodiversity surveys for True Crabs and Wracks - 20 minute timed searches recorded whenever a species from these groups is present and assigning it an abundance category.

We surveyed sites at varying distances away from a sewage pollution point and asked:

  1. Is species diversity correlated with distance from sewage pollution points?

  2. Is variation in species assemblage composition associated with distance from sewage pollution points?

This was also the first time that we used an amended version of the biodiversity survey protocol, which set a low shore limit to the search area as well as the high shore limit. This low shore limit needed to be accessible during surveys and therefore all surveys were conducted at low tide during a spring tide period.

Loading in the data. There are three seperate data sources required for this analysis:

  • The biodiversity survey data (provide as two spreadsheets: “survey_details_jun23.csv” & “species_records_jun_23.csv”)

  • The survey area data. GIS polygons for the survey areas - a SpatialPolygonsDataFrame called RPPsurveysiteGIS (file = “RPPsurveysiteGIS”)

  • The sewage pollution point location data. Provide by the River’s Trust (“RT_dat_Cornwall_Ply.csv”)

#Survey data
species_records <- read.csv("species_records_jun_23.csv")
#removed records listed as 'Unknown' or 'Unlisted' - these are all selfies of surveyors
species_records <- subset(species_records, !grepl("Un", Scientific.name))

survey_details <- read.csv("survey_details_jun23.csv")

#Survey site GIS data
library(rgdal) #GIS package
## Warning: package 'rgdal' was built under R version 4.2.2
## Warning: package 'sp' was built under R version 4.2.2
load("RPPsurveysiteGIS")

#Sewage pollution point data
sewage_point_dat <- read.csv("RT_dat_Cornwall_Ply.csv")

Checking out the survey sites and sewage pollution points

Let’s map out our study areas:

#Falmouth

#get names of Falmouth and Plymouth survey sites with data
survey_sites <- names(table(survey_details$Site))

#remove duplicated sewage points:
sewage_point_dat <- subset(sewage_point_dat, !duplicated(siteNameEA))


fal_sites <- survey_sites[c(2:4,7:10,12,13)]
print(fal_sites)
## [1] "Castle Beach Extreme South" "Castle Beach North"        
## [3] "Castle Beach South"         "Gylly Extreme South"       
## [5] "Gylly North"                "Gylly South"               
## [7] "Gylly steps"                "Tunnel Beach North"        
## [9] "Tunnel Beach South"
ply_sites <- survey_sites[!survey_sites %in% fal_sites]
print(ply_sites)
## [1] "Barn Pool"          "Firestone Bay"      "Firestone Bay West"
## [4] "Mount Batten South"
#subset the GIS data
fal_site_GIS <- subset(RPPsurveysiteGIS, Name %in% fal_sites)
ply_site_GIS <- subset(RPPsurveysiteGIS, Name %in% ply_sites)

#make the survey site maps
library(tmap)
## Warning: package 'tmap' was built under R version 4.2.2
falmap1 <- tm_shape(fal_site_GIS) + 
  tm_borders("dark red") +
  tm_fill("yellow")

plymap1 <- tm_shape(ply_site_GIS)+ 
  tm_borders("dark red") +
  tm_fill("yellow")

tmap_options(check.and.fix = TRUE) #to fix broken polygons

#add the sewage pollution point data
## make spatial points dataframe

sewage_Dat_GIS <- SpatialPointsDataFrame(sewage_point_dat[,c("Longitude","Latitude")], sewage_point_dat, proj4string = RPPsurveysiteGIS@proj4string)

fal_map2 <- (falmap1) +
  tm_shape(sewage_Dat_GIS) +
  tm_symbols(col = "orange") +
  tm_text("siteNameEA", just = "top")

tmap_leaflet(fal_map2)
## Warning: The shape fal_site_GIS is invalid. See sf::st_is_valid
ply_map2 <- (plymap1) +
  tm_shape(sewage_Dat_GIS) +
  tm_symbols(col = "orange") +
  tm_text("siteNameEA", just = "top")

tmap_leaflet(ply_map2)

Species data

Now lets have a quick look at the species data we have collected:

#Total number of surveys
print(nrow(survey_details))
## [1] 32
#breakdown of survey type
print(table(survey_details$Survey.type))
## 
## BiodivTrueCrabs_low_shore_limited    BiodivWracks_low_shore_limited 
##                                15                                17
##True Crab surveys
#overall list of species discovered
tc_surveys <- subset(survey_details, Survey.type == "BiodivTrueCrabs_low_shore_limited")

tc_species_records <- subset(species_records, Survey.ID %in% tc_surveys$Survey.ID)

tc_species <- sort(unique(tc_species_records$Scientific.name))

#six species discovered - how common were they?

tc_abund_cat <- unique(tc_species_records$Abundance)[c(3,1,2,4)] #rearranged in ascending order

tc_abund_cat <- c("Absent", tc_abund_cat) # add absent category

## create table function

sp_abund_sum_tab <- function(surveys, species_records, abund_cats){

blank_dat <- rep(NA,length(abund_cats))
names(blank_dat) <- abund_cats

res <- list()

species <- sort(unique(species_records$Scientific.name))

for (sp in species) {
  
  sp_abund_sum <- blank_dat
  
  for (abund_cat in abund_cats[-1]) {
  
    sp_abund_sum[abund_cat] <- nrow(subset(species_records, Scientific.name == sp & Abundance == abund_cat))
    
      
  }
  
  sp_abund_sum["Absent"] <- nrow(surveys) - sum(sp_abund_sum, na.rm = T)
  
  res[[sp]] <- sp_abund_sum
  
}

res <- do.call("rbind", res)
res <- as.data.frame(res)

return(res)  
    
}

#overall
all_tc_abund_sum <- sp_abund_sum_tab(tc_surveys, tc_species_records, tc_abund_cat)

#just for Falmouth
fal_tc_surv <- subset(tc_surveys, Site %in% fal_sites)
fal_tc_species_rec <- subset(tc_species_records, Survey.ID %in% fal_tc_surv$Survey.ID)
fal_tc_abund_sum <- sp_abund_sum_tab(fal_tc_surv, fal_tc_species_rec, tc_abund_cat)

#just for Plymouth
ply_tc_surv <- subset(tc_surveys, Site %in% ply_sites)
ply_tc_species_rec <- subset(tc_species_records, Survey.ID %in% ply_tc_surv$Survey.ID)
ply_tc_abund_sum <- sp_abund_sum_tab(ply_tc_surv, ply_tc_species_rec, tc_abund_cat)


library(kableExtra)

kable(all_tc_abund_sum, caption = "Abundance categories recorded for all True Crab surveys", "html")%>%
  kable_styling(full_width = F)
Abundance categories recorded for all True Crab surveys
Absent Single Rare <10 Occasional <100 Frequent <1000
Cancer pagurus 4 4 7 0 0
Carcinus maenas 4 5 6 0 0
Maja brachydactyla 14 1 0 0 0
Necora puber 6 4 5 0 0
Xantho hydrophilus 0 0 2 10 3
Xantho pilipes 10 2 3 0 0 
kable(fal_tc_abund_sum, caption = "Abundance categories recorded for Falmouth True Crab surveys", "html")%>%
  kable_styling(full_width = F)
Abundance categories recorded for Falmouth True Crab surveys
Absent Single Rare <10 Occasional <100 Frequent <1000
Cancer pagurus 3 2 4 0 0
Carcinus maenas 4 4 1 0 0
Maja brachydactyla 8 1 0 0 0
Necora puber 4 3 2 0 0
Xantho hydrophilus 0 0 0 7 2
Xantho pilipes 7 1 1 0 0 
kable(ply_tc_abund_sum, caption = "Abundance categories recorded for Plymouth True Crab surveys", "html")%>%
  kable_styling(full_width = F)
Abundance categories recorded for Plymouth True Crab surveys
Absent Single Rare <10 Occasional <100 Frequent <1000
Cancer pagurus 1 2 3 0 0
Carcinus maenas 0 1 5 0 0
Necora puber 2 1 3 0 0
Xantho hydrophilus 0 0 2 3 1
Xantho pilipes 3 1 2 0 0 
#Wracks
#overall list of species discovered
wr_surveys <- subset(survey_details, Survey.type == "BiodivWracks_low_shore_limited")

wr_species_records <- subset(species_records, Survey.ID %in% wr_surveys$Survey.ID)

wr_species <- sort(unique(wr_species_records$Scientific.name))

#nine species discovered - how common were they?

wr_abund_cat <- unique(wr_species_records$Abundance)[c(3,2,4,1,5,6)] #rearranged in ascending order

wr_abund_cat <- c("Absent", wr_abund_cat) # add absent category

#overall
all_wr_abund_sum <- sp_abund_sum_tab(wr_surveys, wr_species_records, wr_abund_cat)

#just for Falmouth
fal_wr_surv <- subset(wr_surveys, Site %in% fal_sites)
fal_wr_species_rec <- subset(wr_species_records, Survey.ID %in% fal_wr_surv$Survey.ID)
fal_wr_abund_sum <- sp_abund_sum_tab(fal_wr_surv, fal_wr_species_rec, wr_abund_cat)

#just for Plymouth
ply_wr_surv <- subset(wr_surveys, Site %in% ply_sites)
ply_wr_species_rec <- subset(wr_species_records, Survey.ID %in% ply_wr_surv$Survey.ID)
ply_wr_abund_sum <- sp_abund_sum_tab(ply_wr_surv, ply_wr_species_rec, wr_abund_cat)

library(kableExtra)

kable(all_tc_abund_sum, caption = "Abundance categories recorded for all True Crab surveys", "html")%>%
  kable_styling(full_width = F)
Abundance categories recorded for all True Crab surveys
Absent Single Rare <10 Occasional <100 Frequent <1000
Cancer pagurus 4 4 7 0 0
Carcinus maenas 4 5 6 0 0
Maja brachydactyla 14 1 0 0 0
Necora puber 6 4 5 0 0
Xantho hydrophilus 0 0 2 10 3
Xantho pilipes 10 2 3 0 0 
kable(fal_tc_abund_sum, caption = "Abundance categories recorded for Falmouth True Crab surveys", "html")%>%
  kable_styling(full_width = F)
Abundance categories recorded for Falmouth True Crab surveys
Absent Single Rare <10 Occasional <100 Frequent <1000
Cancer pagurus 3 2 4 0 0
Carcinus maenas 4 4 1 0 0
Maja brachydactyla 8 1 0 0 0
Necora puber 4 3 2 0 0
Xantho hydrophilus 0 0 0 7 2
Xantho pilipes 7 1 1 0 0 
kable(ply_tc_abund_sum, caption = "Abundance categories recorded for Plymouth True Crab surveys", "html")%>%
  kable_styling(full_width = F)
Abundance categories recorded for Plymouth True Crab surveys
Absent Single Rare <10 Occasional <100 Frequent <1000
Cancer pagurus 1 2 3 0 0
Carcinus maenas 0 1 5 0 0
Necora puber 2 1 3 0 0
Xantho hydrophilus 0 0 2 3 1
Xantho pilipes 3 1 2 0 0

calculating distances between survey sites and sewage pollution points

Before we can beginning looking at our research questions, we need to calculate the distances between our survey sites and sewage pollution points. Ideally this should be calculated as ‘distance by sea’ rather than ‘as the crow flies’ but, looking at the maps, these distances might be similar and, since calcualting distance by sea would be rather complicated, we’ll leave that for now.

In Falmouth, there is a sewage pollution point to the east of our survey sites, around the peninsula called Pendennis Point. The water dynamics in the area may mean that this point is not relevant our our survey sites and, for the sake of this analysis, we will ignore this point and focus solely on the Quenn Mary Gardens Pumping Station. N.B. I have not looked at the hydrology for the area and we may want to look again at this in the future.

In Plymouth, our main sewage pollution point of interest is the Eastern Kings Pumping Station near Devil’s Point. However, there are a number of other sewage outlets close to our survey sites. For this analysis I will not exclude any of Plymouth sewage points and calculate the distance from the centre of each survey area to it’s nearest sewage point.

#coordinates of surveys sites
## Falmouth
fal_co <- coordinates(fal_site_GIS)
colnames(fal_co) <- c("X","Y")
rownames(fal_co) <- fal_site_GIS@data$Name

### Add Falmouth Sewage Point
fal_sew_co <- subset(sewage_point_dat, siteNameEA == "QUEEN MARY GARDENS PUMPING STATION")[,c("Longitude","Latitude")]
colnames(fal_sew_co) <- colnames(fal_co)
fal_co <- rbind(fal_co, fal_sew_co)
rownames(fal_co)[-(1:length(fal_site_GIS))] <- subset(sewage_point_dat, siteNameEA == "QUEEN MARY GARDENS PUMPING STATION")$siteNameEA

### all distances
fal_dist <- spDists(as.matrix(fal_co), longlat = T)
diag(fal_dist) <- NA
colnames(fal_dist) <- rownames(fal_co)
rownames(fal_dist) <- rownames(fal_co)

fal_dist_to_sew <- fal_dist[,"QUEEN MARY GARDENS PUMPING STATION"]

#add to GIS data
fal_site_GIS@data$dist_to_sew <- fal_dist_to_sew[fal_site_GIS@data$Name]

#plot
falmap3 <- tm_shape(fal_site_GIS) + 
  tm_borders("black") +
  tm_fill("dist_to_sew", palette = "RdYlGn", title = "Distance to sewage point (km)") +
  tm_shape(sewage_Dat_GIS) +
  tm_symbols(col = "orange") +
  tm_text("siteNameEA", just = "top")

tmap_leaflet(falmap3)
## Warning: The shape fal_site_GIS is invalid. See sf::st_is_valid
#coordinates of surveys sites
## Plymouth
ply_co <- coordinates(ply_site_GIS)
colnames(ply_co) <- c("X","Y")
rownames(ply_co) <- ply_site_GIS@data$Name

### Add plymouth Sewage Point

ply_sew_co <- subset(sewage_point_dat, localAuthority == "Plymouth")[,c("Longitude","Latitude")]
colnames(ply_sew_co) <- colnames(ply_co)
ply_co <- rbind(ply_co, ply_sew_co)

rownames(ply_co)[-(1:length(ply_site_GIS))] <- subset(sewage_point_dat, localAuthority == "Plymouth")$siteNameEA

### all distances
ply_dist <- spDists(as.matrix(ply_co), longlat = T)
diag(ply_dist) <- NA
colnames(ply_dist) <- rownames(ply_co)
rownames(ply_dist) <- rownames(ply_co)

#make matrix of distances to sewage points

sew_dist_mat <- ply_dist[subset(sewage_point_dat, localAuthority == "Plymouth")$siteNameEA, ply_site_GIS@data$Name ]

#nearest point
sew_dist_min <- apply(sew_dist_mat, 2, min)


#add to GIS data
ply_site_GIS@data$dist_to_sew <- sew_dist_min[ply_site_GIS@data$Name]

#plot
plymap3 <- tm_shape(ply_site_GIS) + 
  tm_borders("black") +
  tm_fill("dist_to_sew", palette = "RdYlGn", title = "Distance to sewage point (km)") +
  tm_shape(sewage_Dat_GIS) +
  tm_symbols(col = "orange") +
  tm_text("siteNameEA", just = "top")

tmap_leaflet(plymap3)