Skip to contents

QC report

2024-07-31

QC.Rmd
# PUT THE YEAR AND THE MONTH YOU WANT TO DO QC
YEAR  <- 2024
MONTH <- 7
library(fmr)
library(tidyverse)
key <- "Your FM API key"

Preamble

This document is currently a combination of notes and some ad-hoc code associated with quality control. In the end we expect to have a more succinct document that only deals with the quality control.

The survey table

We (I) expect that the information in the survey table is refecting information about the whereabout and time that an observer was surveying.

survey <- 
  # want all the columns here, hence trim FALSE
  fmr::fm_survey(key, trim = FALSE) |> 
  filter(year(date) == YEAR,
         month(date) == MONTH)

We do expect survey’s to start and end within the same date 

survey |> 
  filter(date(T1) != date(T2)) |> 
  select(site, date, survey_id, status, collector, T1, T2) |> 
  arrange(site, date, collector) |> 
  knitr::kable()
site date survey_id status collector T1 T2

We do not expect survey’s being done on a Saturday or Sunday 

survey |> 
  mutate(wday = wday(date, label = TRUE, locale = "en_GB.utf8", week_start = 1)) |> 
  filter(wday %in% c("Sat", "Sun")) |> 
  select(site, date, wday, survey_id, status, collector) |> 
  arrange(site, date, collector) |> 
  knitr::kable()
site date wday survey_id status collector
Basseterre Fisheries Complex 2024-07-07 Sun 7777 Completed Rezaine David
Basseterre Fisheries Complex 2024-07-28 Sun 7889 Completed Rezaine David
Dieppe Bay 2024-07-20 Sat 7947 Completed Thrizen Jamas Leader

We do not expect more than one survey entry per observer per sampling day 

d <- 
  survey |> 
  group_by(site, date, collector) |> 
  summarise(n = n(),
            .groups = "drop")
# List of sites, date and collector with more than one survey id
d |> 
  filter(n > 1) |> 
  knitr::kable(caption = "Number of distinct survey_id's by site, date and collector")
Number of distinct survey_id’s by site, date and collector
site date collector n
Palmetto Point 2024-07-16 Tracy-Ann Victoria Gaskin 2 
# Get a list of the survey id's behind these records
d |> 
  filter(n > 1) |> 
  left_join(survey) |> 
  select(site:n, survey_id, status, T1, T2) |> 
  knitr::kable(caption = "List of survey_id's")
List of survey_id’s
site date collector n survey_id status T1 T2
Palmetto Point 2024-07-16 Tracy-Ann Victoria Gaskin 2 7853 Planned 2024-07-16 08:00:00 2024-07-16 16:00:00
Palmetto Point 2024-07-16 Tracy-Ann Victoria Gaskin 2 7856 Planned 2024-07-16 08:00:00 2024-07-16 16:00:00

NOTE: Ideally we would like to have one survey_id per landing site irrespective of the number of surveyor’s. I guess in order to do achieve that one need to setup the anticipated survey date beforehand and then each of the observers log into that same survey_id.