# INBO CODING CLUB

25 April 2024

Welcome!

## Share your code snippet

If you want to share your code snippet, copy paste your snippet within a section of three backticks (```):

As an **example**:

library(tidyverse)

(*you can copy paste this example and add your code further down*)

## Yellow sticky notes

No yellow sticky notes online. Put your name + " | " and add a "*" each time you solve a challenge (see below).

## Participants

Name | Challenges
--- | ---
Damiano Oldoni | ***
Ward Langeraert 🐌 | ***
Hans Van Calster | ***
Luc De Bruyn |
Adriaan Seynaeve | *
Pieter Huybrechts | ***
Sanne Govaert | **
Siebe Indestege | *
Tina Tuerlings |
Jolien Van Malderen | *
Lies Teunen |
Marieke Desender|
Emma Cartuyvels| *

## Challenge 1

### Damiano's solution (example)

Copy paste this section to show your solutions.

# dummy code
print("This is how to insert code.")

### Pieter's solution

## Display first 8 rows; display the last 8 rows.
slice_head(obs, n = 8)

## Select columns observationLevel, eventStart, eventEnd and scientificName
slice_tail(obs, n = 8)

## Display the distinct values of observationLevel as a data.frame with one
## column, observationLevel. How many different values of observationLevel are
## there?

distinct(select(obs, observationLevel))

### How to remove observations with observationLevel = "media"? How to remove
filter(obs, observationLevel != "media")

### observations with observationLevel = "media" and empty scientificName?
filter(obs, observationLevel == "media" & !is.na(scientificName))

### Add a new column called is_classified_by_human which is TRUE if
### classificationMethod is equal to "human". Add a new column month with the
### month of the observation (eventStart). Tip: use lubridate's month() and
### year() functions.
mutate(obs,
       is_classified_human = classificationMethod == "human",
       month = lubridate::month(eventStart)) %>%
  glimpse()

## Move the column count after behavior.
relocate(obs, count, .after = behavior) %>%
  glimpse()

## Move all columns starting with "individual" before observationLevel
relocate(obs, starts_with("individual"), .before = observationLevel) %>%
  glimpse()

## Is the following statement true of false: Observations with observationLevel
## = "event" have no mediaID, i.e. mediaID is NA.
setequal(
  filter(obs, observationLevel == "event" & is.na(mediaID)),
  filter(obs, observationLevel == "event")
)

### Hans

## CHALLENGE 1

obs <- readr::read_csv("./data/20240425/20240425_observations.csv")

# Preview
glimpse(obs)

# Display first 8 rows; display the last 8 rows.

head(obs, 8)
tail(obs, 8)

# Select columns observationLevel, eventStart, eventEnd and scientificName

obs %>%
  select(observationLevel, eventStart, eventEnd, scientificName)

# Display the distinct values of observationLevel as a data.frame with one column, observationLevel.

obs %>%
  distinct(observationLevel)

# How many different values of observationLevel are there?
obs %>%
  count(observationLevel)


#   How to remove observations with observationLevel = "media"? How to remove observations with observationLevel = "media" and empty scientificName?

obs %>%
  filter(observationLevel != "media")

# How to remove observations with observationLevel = "media" and empty scientificName?

obs %>%
  filter(
    observationLevel != "media",
    !is.na(scientificName)
  )


#   Add a new column called is_classified_by_human which is TRUE if classificationMethod is equal to "human". Add a new column month with the month of the observation (eventStart). Tip: use lubridate's month() and year() functions.


obs %>%
  mutate(
    is_classified_by_human = classificationMethod == "human",
    month = lubridate::month(eventStart),
    year = lubridate::year(eventStart)
  )

# Move the column count after behavior.

obs %>%
  relocate(count, .before = behavior)

# Move all columns starting with "individual" before observationLevel

obs %>%
  relocate(
    starts_with("individual"),
    .before = observationLevel
  )


# Is the following statement true of false: Observations with observationLevel = "event" have no mediaID, i.e. mediaID is NA.

obs %>%
  filter(
    observationLevel == "event"
  ) %>%
  select(mediaID) %>%
  pull() %>%
  is.na() %>%
  all()

### Ward's solution

## CHALLENGE 1
# load data
obs <- readr::read_csv("./data/20240425/20240425_observations.csv")

# preview data
glimpse(obs)

# 1. Display first 8 rows; display the last 8 rows.
slice_head(obs, n = 8) # first
slice_tail(obs, n = 8) # last

# 2. Select columns observationLevel, eventStart, eventEnd and scientificName
obs %>%
  select("observationLevel", "eventStart", "eventEnd", "scientificName")

# 3. Display the distinct values of observationLevel as a data.frame with one
# column, observationLevel. How many different values of observationLevel are
# there?
obs %>%
  distinct(.data$observationLevel)

# how many?
obs %>%
  distinct(.data$observationLevel) %>%
  nrow()

# 4. How to remove observations with observationLevel = "media"? How to remove
# observations with observationLevel = "media" and empty scientificName?
obs %>%
  filter(.data$observationLevel != "media")

obs %>%
  filter(.data$observationLevel != "media",
         !is.na(.data$scientificName))

# 5. Add a new column called is_classified_by_human which is TRUE if
# classificationMethod is equal to "human". Add a new column month with the
# month of the observation (eventStart). Tip: use lubridate's month() and year()
# functions.

obs %>%
  mutate(is_classified_by_human = .data$classificationMethod == "human",
         month = month(.data$eventStart))

# 6. Move the column count after behavior.
obs %>%
  relocate("count", .after = "behavior")

# 7. Move all columns starting with "individual." before observationLevel
obs %>%
  relocate(starts_with("individual"), .before = "observationLevel")

# 8. Is the following statement true of false: Observations with
# observationLevel = "event" have no mediaID, i.e. mediaID is NA.
media_events <- obs %>%
  filter(.data$observationLevel == "event") %>%
  pull("mediaID") %>%
  # are they all NA?
  is.na() %>%
  all()

### Emma's solution

# 1
obs %>% slice_head(n = 8)
obs %>% slice_tail(n = 8)

# 2
obs %>% select(observationLevel, eventStart, eventEnd, scientificName)

# 3
obs %>% distinct(observationLevel)

# 4 
obs %>% filter(observationLevel != "media")
obs %>% filter(observationLevel != "media" &
                 !is.na(scientificName))

# 5
obs %>% 
  mutate(is_classified_by_human = ifelse(
    classificationMethod == "human",
    TRUE,
    FALSE
  ),
  month = month(eventStart))

# 6
obs %>% relocate(count, .after = behavior)

# 7
obs %>% relocate(starts_with("individual"),
                 .before = observationLevel)

# 8
obs %>% 
  select(observationLevel, mediaID) %>%
  filter(observationLevel == "event") %>% 
  distinct()

### Siebe's solution

# 1
head(obs,8)
tail(obs,8)

# 2
obs %>% select(observationLevel,eventStart,eventEnd,scientificName)

# 3
obs %>% distinct(observationLevel)

# 4
obs %>% filter(observationLevel != "media")
obs %>% filter(observationLevel != "media" & is.na(scientificName))

# 5
obs <- obs %>% mutate(is_classified_by_human = 
                 case_when(classificationMethod == "human" ~ "TRUE",
                           classificationMethod != "human" ~ "FALSE"),
                 month = month(eventStart)
)
# 6
obs <- obs %>% relocate(count, .after = behavior)

# 7 taxon = individual
obs <- obs %>% relocate(individualID,individualPositionRadius,
                         individualPositionAngle,individualSpeed, 
                         .before = observationLevel
)
# 8
obs %>% filter(observationLevel == "event") %>% distinct(mediaID) #statement = true

### Sanne's solution

# 1.1
obs %>% slice_head(n = 8)
obs %>% slice_tail(n = 8)

# 1.2
obs %>% select(observationLevel, eventStart, eventEnd)

# 1.3
obs %>% distinct(observationLevel)
obs %>%
  distinct(observationLevel) %>%
  nrow()

# 1.4
obs %>% filter(observationLevel != "media")
obs %>%
  filter(observationLevel != "media" & !is.na(scientificName)) 

# 1.5
obs %>%
  mutate(
    is_classified_by =
      if_else(classificationMethod == "human", TRUE, FALSE),
    month = paste(month(eventStart), year(eventStart)),
    )

# 1.6
obs %>%
  relocate(count, .after = behavior)

# 1.7
obs %>%
  relocate(starts_with("individual"), .before = observationLevel) %>%
  names()

# 1.8
obs %>%
  filter(observationLevel == "event") %>%
  distinct(mediaID)

## Challenge 2

Hans

## CHALLENGE 2 - Summaries

# How many observations are there per deploymentID? Show it as a dataframe with two columns: deploymentID and n_obs.

obs %>%
  count(deploymentID, name = "n_obs")

# How many observations are there per deploymentID and observationLevel?

obs %>%
  count(deploymentID, observationLevel, name = "n_obs")


#   Order the previous output per n_obs (descending order).
obs %>%
  count(deploymentID, observationLevel, name = "n_obs") %>%
  arrange(desc(n_obs))

# Create a summary dataframe with the number of media-based observations (observationLevel = "media"), the number of event-based observations (observationLevel = "event") and the total number of observations per deploymentID. Notice that observationLevel is always filled in. The result should be a dataframe with 4 columns: deploymentID, n_obs_event, n_obs_media and n_obs. Tip: one way to solve this is by combining dplyr with tidyr.

obs %>%
  count(deploymentID, observationLevel, name = "n_obs") %>%
  pivot_wider(
    id_cols = deploymentID,
    names_from = observationLevel,
    names_prefix = "n_obs_",
    values_from = n_obs) %>%
  mutate(
    n_obs = n_obs_event + n_obs_media
  )

#Pieter

obs %>% 
  group_by(deploymentID) %>% 
  summarise(n_obs_media = sum(observationLevel == "media"),
            n_obs_event = sum(observationLevel == "event"),
            n_obs = n())
            

# For each month, year and deploymentID, return the eventStart of the first and the last event-based observation (observationLevel = event) if there are 3 or more event-based observations. Call these two columns first_event_obs and last_event_obs respectively.

obs %>%
  mutate(
    month = lubridate::month(eventStart),
    year = lubridate::year(eventStart)
  ) %>%
  group_by(
    deploymentID, month, year
  ) %>%
  filter(
    observationLevel == "event",
    n() >= 3
  ) %>%
  summarise(
    first_event_obs = first(eventStart, order_by = eventStart),
    last_event_obs = last(eventStart, order_by = eventStart),
    .groups = "drop"
  )

# How can you return the same dataframe as in exercice 5 but now limiting us to the months with the highest number of event-based observations for each year and deploymentID?

obs %>%
  mutate(
    month = lubridate::month(eventStart),
    year = lubridate::year(eventStart)
  ) %>%
  group_by(
    deploymentID, month, year
  ) %>%
  mutate(
    n_obs = n()
  ) %>%
  group_by(
    deploymentID, year
  ) %>%
  filter(
    observationLevel == "event",
    n_obs == max(n_obs)
  ) %>%
  group_by(
    deploymentID, month, year
  ) %>%
  summarise(
    first_event_obs = first(eventStart, order_by = eventStart),
    last_event_obs = last(eventStart, order_by = eventStart),
    .groups = "drop"
  )

### Pieter's solution

### How many observations are there per deploymentID? Show it as a dataframe
### with two columns: deploymentID and n_obs.

obs %>%
  group_by(deploymentID) %>%
  tally(name = "n_obs")

count(obs, deploymentID, name = "n_obs")

### How many observations are there per deploymentID and observationLevel?
obs %>%
  group_by(deploymentID, observationLevel) %>%
  tally()

count(obs, deploymentID, observationLevel)

### Order the previous output per n_obs (descending order).

count(obs, deploymentID, observationLevel) %>% 
  group_by(deploymentID) %>% 
  summarise(n_obs = sum(n)) 

obs %>%
  group_by(deploymentID, observationLevel) %>%
  tally(name = "n_obs") %>%
  arrange(-n_obs)

###Create a summary dataframe with the number of media-based observations
###(observationLevel = "media"), the number of event-based observations
###(observationLevel = "event") and the total number of observations per
###deploymentID. Notice that observationLevel is always filled in. The result
###should be a dataframe with 4 columns: deploymentID, n_obs_event, n_obs_media
###and n_obs. Tip: one way to solve this is by combining dplyr with tidyr.

obs %>%
  mutate(is_media = observationLevel == "media",
         is_event = observationLevel == "event") %>%
  group_by(deploymentID) %>%
  summarise(n_obs_media = sum(is_media),
            n_obs_event = sum(is_event),
            n_obs = n())
            
# For each month, year and deploymentID, return the eventStart of the first and
# the last event-based observation (observationLevel = event) if there are 3 or
# more event-based observations. Call these two columns first_event_obs and
# last_event_obs respectively.

obs_first_last <-
  obs %>%
  filter(observationLevel == "event") %>%
  add_count(deploymentID) %>%
  filter(n >= 3) %>%
  mutate(year = lubridate::year(eventStart),
         month = lubridate::month(eventStart)) %>%
  group_by(deploymentID, year, month) %>%
  summarise(
    first_event_obs = head(eventStart, 1),
    last_event_obs = tail(eventStart, 1)
  )

obs_first_last
# How can you return the same dataframe as in exercice 5 but now limiting us to
# the months with the highest number of event-based observations for each year
# and deploymentID?

obs %>%
  filter(observationLevel == "event") %>%
  add_count(deploymentID) %>%
  filter(n >= 3) %>%
  mutate(year = lubridate::year(eventStart),
         month = lubridate::month(eventStart)) %>%
  group_by(deploymentID, year) %>%
  filter(n == max(n)) %>%
  group_by(deploymentID, year, month, .add = FALSE) %>%
  summarise(
    first_event_obs = head(eventStart, 1),
    last_event_obs = tail(eventStart, 1)
  )            

### Ward's solution

# 1. How many observations are there per deploymentID? Show it as a dataframe
# with two columns: deploymentID and n_obs.

obs %>%
  count(.data$deploymentID, name = "n_obs")

# 2. How many observations are there per deploymentID and observationLevel?

count_deploy_obslevel <- obs %>%
  count(.data$deploymentID, .data$observationLevel, name = "n_obs")
count_deploy_obslevel

# 3. Order the previous output per n_obs (descending order).

count_deploy_obslevel %>%
  arrange(desc(.data$n_obs))

# 4. Create a summary dataframe with the number of media-based observations
# (observationLevel = "media"), the number of event-based observations
# (observationLevel = "event") and the total number of observations per
# deploymentID. Notice that observationLevel is always filled in. The result
# should be a dataframe with 4 columns: deploymentID, n_obs_event, n_obs_media
# and n_obs. Tip: one way to solve this is by combining dplyr with tidyr.

obs %>%
  # count number of observations per deployment ID
  count(.data$deploymentID, .data$observationLevel) %>%
  # set to wide format
  pivot_wider(id_cols = "deploymentID",
              names_from = "observationLevel",
              values_from = "n",
              names_prefix = "n_obs_") %>%
  # sum observations across 2 columns
  rowwise() %>%
  mutate(n_obs = sum(c_across(c("n_obs_event", "n_obs_media"))))

# 5. For each month, year and deploymentID, return the eventStart of the first
# and the last event-based observation (observationLevel = event) if there are
# 3 or more event-based observations. Call these two columns first_event_obs and
# last_event_obs respectively.

obs %>%
  # get month and year using lubridate
  mutate(month = month(.data$eventStart),
         year = year(.data$eventStart)) %>%
  # only event-based observations
  filter(.data$observationLevel == "event") %>%
  # count number of observations per month, year, deploymentID
  group_by(.data$month, .data$year, .data$deploymentID) %>%
  mutate(n_obs = n()) %>%
  ungroup() %>%
  # only if there are 3 or more observations
  filter(.data$n_obs >= 3) %>%
  # get first and last observations per month, year, deploymentID
  group_by(.data$month, .data$year, .data$deploymentID) %>%
  summarise(first_event_obs = min(.data$eventStart),
            last_event_obs = max(.data$eventStart),
            .groups = "drop")

# 6. How can you return the same dataframe as in exercice 5 but now limiting us
# to the months with the highest number of event-based observations for each
# year and deploymentID?

obs %>%
  # get month and year using lubridate
  mutate(month = month(.data$eventStart),
         year = year(.data$eventStart)) %>%
  # only event-based observations
  filter(.data$observationLevel == "event") %>%
  # count number of observations per month, year, deploymentID
  group_by(.data$month, .data$year, .data$deploymentID) %>%
  mutate(n_obs = n()) %>%
  ungroup() %>%
  # only if there are 3 or more observations
  filter(.data$n_obs >= 3) %>%
  # keep only the months with the highest number of event-based observations
  group_by(.data$year, .data$deploymentID) %>%
  slice_max(.data$n_obs) %>%
  # get first and last observations per month, year, deploymentID
  group_by(.data$month, .data$year, .data$deploymentID) %>%
  summarise(first_event_obs = min(.data$eventStart),
            last_event_obs = max(.data$eventStart),
            .groups = "drop")

### Sanne's solution

# 2.1 How many observations are there per deploymentID? Show it as a dataframe
# with two columns: deploymentID and n_obs.

obs %>%
  count(deploymentID, name = "n_obs")

# 2.2 How many observations are there per deploymentID and observationLevel?

obs %>%
  count(deploymentID, observationLevel)

# 2.3 Order the previous output per n_obs (descending order).

obs %>%
  count(deploymentID, observationLevel, name = "n_obs") %>%
  arrange(desc(n_obs))

# 2.4 Create a summary dataframe with the number of media-based observations
# (observationLevel = "media"), the number of event-based observations
# (observationLevel = "event") and the total number of observations per
# deploymentID. Notice that observationLevel is always filled in. The result
# should be a dataframe with 4 columns: deploymentID, n_obs_event, n_obs_media
# and n_obs. Tip: one way to solve this is by combining dplyr with tidyr.

obs %>%
  count(deploymentID, observationLevel, name = "n_obs") %>%
  summarise(n_obs = n()) %>%
  pivot_wider(
    names_from = observationLevel,
    values_from = n_obs,
    names_glue = "n_obs_{observationLevel}") %>%
  mutate(
    n_obs = n_obs_event + n_obs_media
  )

# 2.5 For each month, year and deploymentID, return the eventStart of the first
# and the last event-based observation (observationLevel = event) if there are 3
# or more event-based observations. Call these two columns first_event_obs and
# last_event_obs respectively.

groupedobs <- obs %>%
  mutate(
    month = month(eventStart),
    year = year(eventStart)
  ) %>%
  group_by(month, year, deploymentID) %>%
  filter(observationLevel == "event" & n() >= 3) 

firstlast <- groupedobs %>%
  summarize(
    first_event_obs = first(eventStart, order_by = eventStart),
    last_event_obs = last(eventStart, order_by = eventStart)
  ) %>% 
  ungroup()

firstlast

# 2.6 How can you return the same dataframe as in exercice 5 but now limiting us
# to the months with the highest number of event-based observations for each year
# and deploymentID?

groupedobs %>%
  count() %>% 
  group_by(year, deploymentID) %>% 
  filter(n == max(n)) %>% 
  ungroup() %>% 
  left_join(firstlast) %>% 
  select(-n)

### Marieke's solutions

chal2_obs4<-obs %>% 
  group_by(deploymentID) %>%
  summarise(., n_obs_media=n_distinct(observationID[observationLevel=="media"]), 
            n_obs_event=n_distinct(observationID[observationLevel=="event"]),
            n_obs=n_distinct(observationID))
 

### Emma's solution

# 5

obs %>%
  filter(observationLevel == "event") %>% 
  mutate(month = month(eventStart),
         year = year(eventStart)) %>%
  mutate(n_obs = n(), .by = c(month, year, deploymentID)) %>% 
  filter(n_obs > 2) %>% 
  summarise(first_event_obs = min(eventStart),
            last_event_obs =  max(eventStart),
            .by = c(month, year, deploymentID)) %>% 
  arrange(year, month)

## CHALLENGE 3 - Two-table verbs

media <- read_csv("./data/20240425/20240425_media.csv", na = "")

# Preview
glimpse(media)


#  How to add the media information stored in media to the observations?

obs %>%
  left_join(
    media,
    by = join_by(deploymentID, mediaID)
  )

#  How are the columns ordered? Are the columns from observations on the left? Try to put them on the right.

media %>%
  right_join(
    obs,
    by = join_by(deploymentID, mediaID)
  )


#  Are there media not in observations, i.e. are there media that are not linked to any observation?

media %>%
  anti_join(
    obs,
    by = join_by(deploymentID, mediaID)
  ) %>%
  nrow() > 0
# NO


#  Some observations have a missing value for column mediaID. Get rid of them while joining.

obs %>%
  inner_join(
    media,
    by = join_by(deploymentID, mediaID)
  )


#  As deploymentID is present in both dataframes, it gets duplicated and the suffixes ".x" and ".y" are added. How to change the suffixes to "_obs" and "_media" while performing exercise 1?

obs %>%
  inner_join(
    media,
    by = join_by(mediaID),
    suffix = c("_obs", "_media")
  )


#  How to add the suffix only for the column deploymentID in media?
obs %>%
  inner_join(
    media,
    by = join_by(mediaID),
    suffix = c("", "_media")
  )

#  How can you avoid having this column twice?
obs %>%
  inner_join(
    media,
    by = join_by(mediaID, deploymentID))

### Pieter's solution

## How to add the media information stored in media to the observations?
left_join(obs, media, by = join_by("mediaID")) %>%
  glimpse()

## How are the columns ordered? Are the columns from observations on the left?
## Try to put them on the right.
right_join(media, obs, by = join_by("mediaID")) %>%
  glimpse()

## Are there media not in observations, i.e. are there media that are not linked
## to any observation?
anti_join(media, obs, by  = join_by("mediaID"))

## Some observations have a missing value for column mediaID. Get rid of them
## while joining.
inner_join(obs, media, by = join_by("mediaID"))

## As deploymentID is present in both dataframes, it gets duplicated and the
## suffixes ".x" and ".y" are added. How to change the suffixes to "_obs" and
## "_media" while performing exercise 1?
left_join(obs,
          media,
          by = join_by("mediaID"),
          suffix = c("_obs", "_media")) %>%
  glimpse()

## How to add the suffix only for the column deploymentID in media?
mutate(media, deploymentID_media = deploymentID, .keep = "unused") %>%
  right_join(obs, by = join_by("mediaID")) %>%
  glimpse()

## How can you avoid to have this column twice?
obs %>%
  left_join(select(media, -deploymentID), by = join_by("mediaID")) %>%
  glimpse()`

### Ward's solution

# load data
media <- read_csv("./data/20240425/20240425_media.csv", na = "")

# preview data
glimpse(media)

# 1. How to add the media information stored in media to the observations?

obs %>%
  full_join(media, by = join_by("mediaID"))

# 2. How are the columns ordered? Are the columns from observations on the left?
# Try to put them on the right.

media %>%
  full_join(obs, by = join_by("mediaID"))

# 3. Are there media not in observations, i.e. are there media that are not
# linked to any observation?

media %>%
  anti_join(obs, by = join_by("mediaID"))

# 4. Some observations have a missing value for column mediaID. Get rid of them
# while joining.

obs %>%
  right_join(media, by = join_by("mediaID"))

# 5. As deploymentID is present in both dataframes, it gets duplicated and the
# suffixes ".x" and ".y" are added. How to change the suffixes to "_obs" and
# "_media" while performing exercise 1?

obs %>%
  full_join(media, by = join_by("mediaID"), suffix = c("_obs", "_media"))

# 6. How to add the suffix only for the column deploymentID in media?

obs %>%
  full_join(media, by = join_by("mediaID"), suffix = c("", "_media"))

# 7. How can you avoid having this column twice?

obs %>%
  full_join(media, by = join_by("mediaID", "deploymentID"))

## Bonus 1

Hans

#After reading 20240425_ias_plants.txt dataset (code provided), make it tidy. Please notice that two vernacular names of the same language are given for some species, e.g. "gewone gunnera;reuzenrabarber". They should be split as two rows.

ias %>%
  pivot_longer(
    cols = c(english_name, dutch_name, french_name),
    names_to = c("language", "drop_me"),
    names_sep = "_",
    values_to = "name") %>%
  select(-drop_me) %>%
  separate_wider_delim(
    cols = name,
    names = c("name1", "name2"),
    delim = ";",
    too_few = "align_start"
  ) %>%
  pivot_longer(
    cols = starts_with("name"),
    names_to = "name_alternative",
    values_to = "name",
    values_drop_na = TRUE
  )

## Bonus 2

Hans

a %>%
  slice(
    order(as.numeric(my_favorite_number_string))
  )
  

## Bonus 3

Hans

result <- si_data %>%
  group_by(year, eea_cell_code) %>%
  summarise(species_list = list(as.character(speciesKey)))  %>%
  mutate(
    accumulated_species_list = accumulate(
      species_list, union, .simplify = FALSE),
    cumulative_n_species = map_dbl(accumulated_species_list, length)
  ) %>%
  ungroup()

library(ggplot2)
result %>%
  ggplot() +
  geom_line(aes(x = year, y = cumulative_n_species,
                group = eea_cell_code),
            alpha = 0.3)