Wildlife Comeback in Europe 2021
Indicators and Assessments Unit (SD, RF)
19/07/2021
Analysis of population and range change in European mammals from Article 17 data
Explore population trends and distribution changes of the Wildlife Comeback in Europe Report species that have records within the Article 17 data.
mammals_list = c(
"Megaptera_novaeangliae",
#"Caretta_caretta", ##Loggerheads
"Pusa_hispida",
"Lutra_lutra",
"Meles_meles",
"Martes_martes",
"Myotis_daubentonii",
"Myotis_nattereri",
"Myotis_emarginatus",
"Alces_alces",
"Bison_bonasus",
"Canis_aureus",
"Canis_lupus",
"Capra_ibex",
"Capra_pyrenaica",
"Capreolus_capreolus",
"Castor_fiber",
"Cervus_elaphus",
"Gulo_gulo",
"Halichoerus_grypus",
"Lynx_lynx",
"Lynx_pardinus",
"Phoca_vitulina",
"Rupicapra_pyrenaica",
"Rupicapra_rupicapra",
"Sus_scrofa",
"Ursus_arctos")
mammals_list = gsub("_", " ", mammals_list)List of mammals for analysis
mammals_list## [1] "Megaptera novaeangliae" "Pusa hispida" "Lutra lutra"
## [4] "Meles meles" "Martes martes" "Myotis daubentonii"
## [7] "Myotis nattereri" "Myotis emarginatus" "Alces alces"
## [10] "Bison bonasus" "Canis aureus" "Canis lupus"
## [13] "Capra ibex" "Capra pyrenaica" "Capreolus capreolus"
## [16] "Castor fiber" "Cervus elaphus" "Gulo gulo"
## [19] "Halichoerus grypus" "Lynx lynx" "Lynx pardinus"
## [22] "Phoca vitulina" "Rupicapra pyrenaica" "Rupicapra rupicapra"
## [25] "Sus scrofa" "Ursus arctos"How many do we match in article 17 data?
#### *** 2020 species data
#mammal_species <- fread("Article17_2020_dataset/Article17_2020_speciesEUassessment.csv", na.strings = 'NULL')
mammal_species <- fread("Article17_2020_dataset/Article17_2020_data_species_regions.csv", na.strings = 'NULL')
#write.csv(bird_species,"bird_species_check.csv")
#intersect(mammal_species$assessment_speciesname, mammals_list)
intersect(mammal_species$speciesname, mammals_list)## [1] "Myotis daubentonii" "Myotis emarginatus" "Myotis nattereri"
## [4] "Castor fiber" "Canis lupus" "Canis aureus"
## [7] "Ursus arctos" "Lutra lutra" "Martes martes"
## [10] "Lynx lynx" "Rupicapra rupicapra" "Capra ibex"
## [13] "Halichoerus grypus" "Phoca vitulina" "Bison bonasus"
## [16] "Megaptera novaeangliae" "Lynx pardinus" "Capra pyrenaica"
## [19] "Rupicapra pyrenaica" "Gulo gulo" "Pusa hispida"setdiff(mammals_list, mammal_species$speciesname)## [1] "Meles meles" "Alces alces" "Capreolus capreolus"
## [4] "Cervus elaphus" "Sus scrofa"## 21 of 26 match****
#Meles meles
#Sus scrofa
#Alces alces
#Capreolus capreolus
#Cervus elaphus
## I think these are missing as they're not Article 17 species (check?)
# Restrict to those in list
# **
# ** Need to check matches as losing some
# **
mammal_species_short <- mammal_species[mammal_species$speciesname %in% mammals_list, ]
#names(bird_species_short)[names(bird_species_short) == "ï..specieshash"] <- "specieshash"
# Get the species codes (unique for each species)
speciescode <- unique(mammal_species$speciescode)Which species have population trend data available?
# Population trend vs species
table(mammal_species_short$speciesname, mammal_species_short$population_trend)##
## D I S U Unk
## Bison bonasus 2 0 3 0 0 1
## Canis aureus 0 0 11 4 0 5
## Canis lupus 5 1 23 13 2 1
## Capra ibex 2 1 3 1 0 0
## Capra pyrenaica 0 0 6 0 0 0
## Castor fiber 0 1 31 4 0 0
## Gulo gulo 0 0 3 1 0 0
## Halichoerus grypus 1 0 13 2 1 2
## Lutra lutra 1 0 25 21 1 6
## Lynx lynx 2 4 5 13 1 8
## Lynx pardinus 0 0 2 0 0 0
## Martes martes 0 0 8 22 3 20
## Megaptera novaeangliae 2 0 0 0 1 10
## Myotis daubentonii 0 4 4 24 7 11
## Myotis emarginatus 1 5 12 16 1 7
## Myotis nattereri 0 0 9 18 9 14
## Phoca vitulina 1 0 9 1 1 3
## Pusa hispida 1 0 0 0 0 1
## Rupicapra pyrenaica 0 0 3 3 0 0
## Rupicapra rupicapra 3 0 1 10 0 3
## Ursus arctos 0 3 12 10 1 0Which species have long-term population trend data available?
# Population trend vs species
table(mammal_species_short$speciesname, mammal_species_short$population_trend_long)##
## D I S U Unk
## Bison bonasus 4 0 2 0 0 0
## Canis aureus 16 0 3 1 0 0
## Canis lupus 32 1 9 2 1 0
## Capra ibex 4 0 3 0 0 0
## Capra pyrenaica 4 0 2 0 0 0
## Castor fiber 21 0 15 0 0 0
## Gulo gulo 4 0 0 0 0 0
## Halichoerus grypus 12 0 7 0 0 0
## Lutra lutra 36 1 16 1 0 0
## Lynx lynx 29 1 1 2 0 0
## Lynx pardinus 1 0 1 0 0 0
## Martes martes 40 0 6 6 0 1
## Megaptera novaeangliae 9 0 0 0 0 4
## Myotis daubentonii 32 1 0 7 2 8
## Myotis emarginatus 30 2 6 1 0 3
## Myotis nattereri 35 1 4 5 0 5
## Phoca vitulina 8 0 6 0 1 0
## Pusa hispida 2 0 0 0 0 0
## Rupicapra pyrenaica 2 0 2 2 0 0
## Rupicapra rupicapra 13 0 3 1 0 0
## Ursus arctos 19 0 6 1 0 0Which species have regional range change data available?
# Population trend vs species
table(mammal_species_short$speciesname, mammal_species_short$range_trend)##
## D I S U Unk
## Bison bonasus 0 0 3 0 0 3
## Canis aureus 0 0 14 4 0 2
## Canis lupus 4 0 23 15 2 1
## Capra ibex 2 0 1 4 0 0
## Capra pyrenaica 0 0 2 4 0 0
## Castor fiber 0 0 32 4 0 0
## Gulo gulo 0 0 2 2 0 0
## Halichoerus grypus 3 0 1 15 0 0
## Lutra lutra 1 0 25 25 0 3
## Lynx lynx 2 3 6 15 2 5
## Lynx pardinus 0 0 2 0 0 0
## Martes martes 0 0 8 37 0 8
## Megaptera novaeangliae 4 0 0 2 0 7
## Myotis daubentonii 0 0 3 42 0 5
## Myotis emarginatus 1 1 2 32 0 6
## Myotis nattereri 0 1 4 36 3 6
## Phoca vitulina 2 0 3 9 0 1
## Pusa hispida 2 0 0 0 0 0
## Rupicapra pyrenaica 0 0 2 4 0 0
## Rupicapra rupicapra 3 0 1 12 0 1
## Ursus arctos 0 2 10 14 0 0Which species have regional long-term range change data available?
# Population trend vs species
table(mammal_species_short$speciesname, mammal_species_short$range_trend_long)##
## D I S U Unk
## Bison bonasus 4 0 2 0 0 0
## Canis aureus 16 0 4 0 0 0
## Canis lupus 30 0 11 3 0 1
## Capra ibex 4 0 1 2 0 0
## Capra pyrenaica 3 0 2 1 0 0
## Castor fiber 21 0 15 0 0 0
## Gulo gulo 3 0 1 0 0 0
## Halichoerus grypus 14 0 3 1 1 0
## Lutra lutra 35 1 16 2 0 0
## Lynx lynx 28 1 2 2 0 0
## Lynx pardinus 1 0 1 0 0 0
## Martes martes 40 0 4 9 0 0
## Megaptera novaeangliae 7 0 0 0 1 5
## Myotis daubentonii 34 0 0 12 0 4
## Myotis emarginatus 29 1 4 3 0 5
## Myotis nattereri 37 0 2 7 0 4
## Phoca vitulina 10 0 1 3 1 0
## Pusa hispida 2 0 0 0 0 0
## Rupicapra pyrenaica 3 0 1 2 0 0
## Rupicapra rupicapra 12 0 2 3 0 0
## Ursus arctos 19 1 5 1 0 0Which species have regional habitat change data available?
# Population trend vs species
table(mammal_species_short$speciesname, mammal_species_short$habitat_trend)##
## D I S U Unk
## Bison bonasus 0 0 1 4 0 1
## Canis aureus 2 0 6 12 0 0
## Canis lupus 8 1 10 26 0 0
## Capra ibex 2 0 0 5 0 0
## Capra pyrenaica 0 0 0 6 0 0
## Castor fiber 0 0 14 20 1 1
## Gulo gulo 0 0 0 4 0 0
## Halichoerus grypus 2 0 0 13 1 3
## Lutra lutra 1 1 4 41 1 6
## Lynx lynx 2 1 1 19 1 9
## Lynx pardinus 0 0 0 1 1 0
## Martes martes 0 1 2 29 5 16
## Megaptera novaeangliae 4 0 0 1 0 8
## Myotis daubentonii 0 5 0 29 0 16
## Myotis emarginatus 1 12 0 20 1 8
## Myotis nattereri 0 2 2 29 3 14
## Phoca vitulina 2 0 1 8 0 4
## Pusa hispida 1 0 0 0 0 1
## Rupicapra pyrenaica 0 0 0 3 0 3
## Rupicapra rupicapra 3 0 0 8 0 6
## Ursus arctos 1 0 2 22 0 1Which species have regional long-term habitat change data available?
# Population trend vs species
table(mammal_species_short$speciesname, mammal_species_short$habitat_trend_long)##
## D I S U Unk
## Bison bonasus 4 0 1 1 0 0
## Canis aureus 18 0 0 2 0 0
## Canis lupus 35 0 4 6 0 0
## Capra ibex 5 0 0 1 1 0
## Capra pyrenaica 3 1 2 0 0 0
## Castor fiber 26 0 2 4 0 4
## Gulo gulo 4 0 0 0 0 0
## Halichoerus grypus 15 0 0 3 1 0
## Lutra lutra 39 1 6 6 1 1
## Lynx lynx 30 0 1 2 0 0
## Lynx pardinus 2 0 0 0 0 0
## Martes martes 41 0 3 9 0 0
## Megaptera novaeangliae 9 0 0 0 1 3
## Myotis daubentonii 34 1 0 8 1 6
## Myotis emarginatus 32 3 0 4 0 3
## Myotis nattereri 38 1 0 5 0 6
## Phoca vitulina 8 0 1 5 1 0
## Pusa hispida 2 0 0 0 0 0
## Rupicapra pyrenaica 3 0 0 2 0 1
## Rupicapra rupicapra 13 0 1 0 0 3
## Ursus arctos 19 0 0 6 0 1##Threat information
Extract and count the number of threats for each species in each country
# Threats info
#### Threats
mammal_threats <- fread("Article17_2020_dataset/Article17_2020_data_pressures_threats.csv")
mammal_threats = subset(mammal_threats, featuretype == "species")
mammal_threats$featurecode = as.numeric(mammal_threats$featurecode) # Some habitat codes become NA if we do this first
# Subset threats for species we care about
mammal_threats = mammal_threats[mammal_threats$featurecode %in% speciescode, ]
# Filter out 'no-pressure' codes
mammal_threats = subset(mammal_threats, pressurecode != "Xxt") # Get rid of Xxt
# Get rid of all of those starting with 'X'??
mammal_threats = subset(mammal_threats, !startsWith(mammal_threats$pressurecode, "X"))
mammal_threats = subset(mammal_threats, !is.na(mammal_threats$pressurecode))
mammal_threats_both = mammal_threats %>%
group_by(country, region, featurename, featurecode, type) %>%
summarise(
n = n_distinct(pressurecode),
n_hl = n_distinct(substring(pressurecode, 1, 1)) #n unique codes using only first character
)## `summarise()` has grouped output by 'country', 'region', 'featurename', 'featurecode'. You can override using the `.groups` argument.# Convert this new data to four columns (n_t, n_p, n_hl_t, n_hl_p)
mammal_threats_n = tidyr::pivot_wider(mammal_threats_both,
id_cols = country:featurecode,
names_from="type",
values_from = c("n", "n_hl"),
values_fill = 0) # Fill missing counts with 0
# Give them better names
mammal_threats_n = rename(mammal_threats_n,
n_threat = n_t,
n_pressure = n_p,
n_hl_threat = n_hl_t,
n_hl_pressure = n_hl_p)
mammal_threats_n## # A tibble: 7,311 x 8
## # Groups: country, region, featurename, featurecode [7,311]
## country region featurename featurecode n_pressure n_threat n_hl_pressure
## <chr> <chr> <chr> <dbl> <int> <int> <int>
## 1 AT ALP Acipenser ruthe… 2487 2 2 2
## 2 AT ALP Adenophora lili… 4068 6 6 5
## 3 AT ALP Alburnus mento 5289 4 4 3
## 4 AT ALP Anisus vorticul… 4056 3 4 3
## 5 AT ALP Apium repens 1614 2 2 1
## 6 AT ALP Aquilegia alpina 1480 2 2 2
## 7 AT ALP Arnica montana 1762 6 6 4
## 8 AT ALP Artemisia genipi 1764 2 2 2
## 9 AT ALP Aspius aspius 1130 2 2 2
## 10 AT ALP Asplenium adult… 4066 3 3 3
## # … with 7,301 more rows, and 1 more variable: n_hl_threat <int># Couple of quick plots to look at relationship between n_threat and n_hl_threat
ggplot(mammal_threats_n, aes(x = n_threat, y = n_hl_threat)) +
geom_point(alpha = 0.05, size = 5) +
ggtitle("Number of unique threats vs. number of unique 'high-level' threats") +
xlab("Number of unique threats") +
ylab("Number of unique high-level threats") +
ggpubr::theme_classic2()## Registered S3 methods overwritten by 'car':
## method from
## influence.merMod lme4
## cooks.distance.influence.merMod lme4
## dfbeta.influence.merMod lme4
## dfbetas.influence.merMod lme4
ggplot(mammal_threats_n, aes(x = n_pressure, y = n_hl_pressure)) +
geom_point(alpha = 0.05, size = 5) +
ggtitle("Number of unique pressures vs. number of unique 'high-level' pressures") +
xlab("Number of unique pressures") +
ylab("Number of unique high-level pressures") +
ggpubr::theme_classic2()
# Measures
mammal_measures <- fread("Article17_2020_dataset/Article17_2020_data_measures.csv", na.strings = 'NULL')
mammal_measures$featurecode = as.numeric(mammal_measures$featurecode) # Some habitat codes become NA## Warning: NAs introduced by coercion# Subset measures for species we care about
mammal_measures <- mammal_measures[mammal_measures$featurecode %in% speciescode, ]
mammal_measures = subset(mammal_measures, featuretype == "species")
# ** Do we need to filter out 'no-measure' codes?
#mammal_threats = subset(mammal_threats, measurecode != "Xxt") ?
mammal_measures_n = mammal_measures %>%
group_by(country, region, featurename, featurecode) %>%
summarise(
n_measure = n_distinct(measurecode),
n_hl_measure = n_distinct(substring(measurecode, 1, 2)) #n unique codes using only first character
) ## `summarise()` has grouped output by 'country', 'region', 'featurename'. You can override using the `.groups` argument.mammal_measures_n## # A tibble: 4,671 x 6
## # Groups: country, region, featurename [4,671]
## country region featurename featurecode n_measure n_hl_measure
## <chr> <chr> <chr> <dbl> <int> <int>
## 1 AT ALP Adenophora lilifolia 4068 1 1
## 2 AT ALP Alburnus mento 5289 2 2
## 3 AT ALP Anisus vorticulus 4056 1 1
## 4 AT ALP Apium repens 1614 1 1
## 5 AT ALP Aspius aspius 1130 2 2
## 6 AT ALP Astacus astacus 1091 7 5
## 7 AT ALP Austropotamobius pallipes 1092 1 1
## 8 AT ALP Austropotamobius torrentium 1093 6 4
## 9 AT ALP Barbastella barbastellus 1308 2 1
## 10 AT ALP Barbus barbus 5085 3 3
## # … with 4,661 more rows# Couple of quick plots to look at relationship between n_threat and n_hl_threat
ggplot(mammal_measures_n, aes(x = n_measure, y = n_hl_measure)) +
geom_point(alpha = 0.05, size = 5) +
ggtitle("Number of unique measures vs. number of unique 'high-level' measures (mammals)") +
xlab("Number of unique threats") +
ylab("Number of unique high-level threats") +
ggpubr::theme_classic2()
##Combine data
# Trend data - get columns from bird species table that we want to use
mammal_trend_data = mammal_species_short %>% select(country, region, speciescode, speciesname,
population_trend,
population_trend_long,
range_trend,
range_trend_long,
habitat_trend,
habitat_trend_long)
mammal_trend_data = rename(mammal_trend_data,
featurecode = speciescode,
featurename = speciesname)
# All together now!
# Combined measure data with threat data
full_dataB = merge(mammal_trend_data, mammal_threats_n, by = c('region', 'country', 'featurename', 'featurecode'), all.x =TRUE)
# Combine those with trend data
full_data = merge(full_dataB, mammal_measures_n, by=c('region', 'country', 'featurename', 'featurecode'), all.x =TRUE)How many species do we end up with?
length(unique(full_data$featurename))## [1] 21How many country records (threats/measures) do we have for each species?
table(full_data$featurename)##
## Bison bonasus Canis aureus Canis lupus
## 6 20 45
## Capra ibex Capra pyrenaica Castor fiber
## 7 6 36
## Gulo gulo Halichoerus grypus Lutra lutra
## 4 19 54
## Lynx lynx Lynx pardinus Martes martes
## 33 2 53
## Megaptera novaeangliae Myotis daubentonii Myotis emarginatus
## 13 50 42
## Myotis nattereri Phoca vitulina Pusa hispida
## 50 15 2
## Rupicapra pyrenaica Rupicapra rupicapra Ursus arctos
## 6 17 26Combine countries to regions
Population/habitat change data is given per regions, so aggregate n_threat/n_measure to region
# Average threats over countries to regions as that's where we have response data
# model_data = full_data %>%
# group_by(region, featurename, featurecode, population_trend, habitat_trend) %>%
# summarise(
# mean_n_pressure = mean(n_pressure),
# mean_n_threat = mean(n_threat),
# mean_n_hl_pressure = mean(n_hl_pressure),
# mean_n_hl_threat = mean(n_hl_threat),
# mean_n_measure = mean(n_measure),
# mean_n_hl_measure = mean(n_hl_measure),
# )
model_data = full_dataHow many species are there per region?
# How many do we have
addmargins(table(model_data$featurename, model_data$region))##
## ALP ATL BLS BOR CON MATL MBAL MED MMAC MMED PAN STE
## Bison bonasus 3 0 0 0 3 0 0 0 0 0 0 0
## Canis aureus 5 0 2 0 7 0 0 2 0 0 3 1
## Canis lupus 12 6 1 5 13 0 0 6 0 0 2 0
## Capra ibex 5 0 0 0 1 0 0 1 0 0 0 0
## Capra pyrenaica 2 2 0 0 0 0 0 2 0 0 0 0
## Castor fiber 7 6 0 5 11 0 0 2 0 0 4 1
## Gulo gulo 2 0 0 2 0 0 0 0 0 0 0 0
## Halichoerus grypus 0 0 0 0 0 10 8 0 1 0 0 0
## Lutra lutra 13 9 2 5 14 0 0 6 0 0 4 1
## Lynx lynx 11 1 1 5 10 0 0 3 0 0 2 0
## Lynx pardinus 0 0 0 0 0 0 0 2 0 0 0 0
## Martes martes 13 9 1 5 14 0 0 6 0 0 4 1
## Megaptera novaeangliae 0 0 0 0 0 6 0 0 2 5 0 0
## Myotis daubentonii 11 9 1 5 14 0 0 6 0 0 3 1
## Myotis emarginatus 11 7 1 0 12 0 0 7 0 0 3 1
## Myotis nattereri 11 9 1 5 14 0 0 7 0 0 3 0
## Phoca vitulina 0 0 0 0 0 10 4 0 1 0 0 0
## Pusa hispida 0 0 0 0 0 2 0 0 0 0 0 0
## Rupicapra pyrenaica 2 2 0 0 0 0 0 2 0 0 0 0
## Rupicapra rupicapra 7 0 0 0 7 0 0 3 0 0 0 0
## Ursus arctos 12 1 0 4 5 0 0 4 0 0 0 0
## Sum 127 61 10 41 125 28 12 59 4 5 28 6
##
## Sum
## Bison bonasus 6
## Canis aureus 20
## Canis lupus 45
## Capra ibex 7
## Capra pyrenaica 6
## Castor fiber 36
## Gulo gulo 4
## Halichoerus grypus 19
## Lutra lutra 54
## Lynx lynx 33
## Lynx pardinus 2
## Martes martes 53
## Megaptera novaeangliae 13
## Myotis daubentonii 50
## Myotis emarginatus 42
## Myotis nattereri 50
## Phoca vitulina 15
## Pusa hispida 2
## Rupicapra pyrenaica 6
## Rupicapra rupicapra 17
## Ursus arctos 26
## Sum 506First simple model.. do species appear to have more conservation measures in place when there are more threats reported?
model_data_measure = model_data %>%
mutate(population_trend = factor(population_trend)) %>%
filter(population_trend != "") %>% # Subset to only increases or declines
select(region, n_hl_threat, n_hl_measure, population_trend_long, featurename) %>% # Columns for modelling
drop_na() %>% # Get rid of NAs in any of these columns
droplevels()
#*** Including species ('featurename') as a random effect is tricky (throw singularities) as there's only a couple
# of records for some species.
m1_measures_NULL = lmer(n_hl_measure ~ (1|region),
data = model_data_measure)
m1_measuresA = lmer(n_hl_measure ~ population_trend_long + n_hl_threat + (1|region),
data = model_data_measure)
m1_measuresB = lmer(n_hl_measure ~ population_trend_long * n_hl_threat + (1|region),
data = model_data_measure)
# Adding 'country' throws singularities due to sparse data
#summary(m1_measures)
tab_model(m1_measures_NULL, m1_measuresA, m1_measuresB, show.aicc = TRUE)| n_hl_measure | n_hl_measure | n_hl_measure | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Predictors | Estimates | CI | p | Estimates | CI | p | Estimates | CI | p |
| (Intercept) | 3.27 | 2.68 – 3.87 | <0.001 | 1.52 | 0.95 – 2.09 | <0.001 | 1.59 | 0.91 – 2.27 | <0.001 |
| population_trend_long [D] | -0.20 | -1.65 – 1.25 | 0.788 | -2.46 | -5.83 – 0.90 | 0.152 | |||
| population_trend_long [I] | -0.26 | -0.70 – 0.17 | 0.228 | -0.36 | -1.38 – 0.66 | 0.493 | |||
| population_trend_long [S] | -0.46 | -1.69 – 0.77 | 0.466 | -1.65 | -5.02 – 1.72 | 0.338 | |||
| population_trend_long [U] | -1.10 | -3.14 – 0.95 | 0.294 | 3.05 | -2.17 – 8.28 | 0.252 | |||
|
population_trend_long [Unk] |
-0.37 | -1.21 – 0.47 | 0.394 | -0.28 | -2.59 – 2.02 | 0.808 | |||
| n_hl_threat | 0.46 | 0.36 – 0.57 | <0.001 | 0.44 | 0.31 – 0.58 | <0.001 | |||
|
population_trend_long [D] * n_hl_threat |
0.64 | -0.22 – 1.51 | 0.146 | ||||||
|
population_trend_long [I] * n_hl_threat |
0.02 | -0.19 – 0.24 | 0.817 | ||||||
|
population_trend_long [S] * n_hl_threat |
0.22 | -0.34 – 0.78 | 0.448 | ||||||
|
population_trend_long [U] * n_hl_threat |
-1.18 | -2.54 – 0.18 | 0.089 | ||||||
|
population_trend_long [Unk] * n_hl_threat |
-0.04 | -0.72 – 0.65 | 0.920 | ||||||
| Random Effects | |||||||||
| σ2 | 2.64 | 2.10 | 2.09 | ||||||
| τ00 | 0.80 region | 0.25 region | 0.27 region | ||||||
| ICC | 0.23 | 0.10 | 0.11 | ||||||
| N | 12 region | 12 region | 12 region | ||||||
| Observations | 265 | 265 | 265 | ||||||
| Marginal R2 / Conditional R2 | 0.000 / 0.233 | 0.244 / 0.323 | 0.252 / 0.337 | ||||||
| AIC | 1035.007 | 975.498 | 983.129 | ||||||
compare_performance(m1_measures_NULL, m1_measuresA, m1_measuresB)## # Comparison of Model Performance Indices
##
## Name | Model | AIC | BIC | R2 (cond.) | R2 (marg.) | ICC | RMSE | Sigma
## --------------------------------------------------------------------------------------------------
## m1_measures_NULL | lmerMod | 1034.915 | 1045.654 | 0.233 | 0.000 | 0.233 | 1.598 | 1.625
## m1_measuresA | lmerMod | 974.792 | 1007.010 | 0.323 | 0.244 | 0.105 | 1.413 | 1.449
## m1_measuresB | lmerMod | 981.449 | 1031.565 | 0.337 | 0.252 | 0.114 | 1.395 | 1.446#
# library(rstanarm)
# sm1_measures_NULL = stan_glmer(n_hl_measure ~ (1|featurename),
# data = model_data_measure, diagnostic_file = file.path(tempdir(), "df0.csv"))
# sm1_measuresA = stan_glmer(n_hl_measure ~ population_trend + n_hl_threat + (1|featurename),
# data = model_data_measure, diagnostic_file = file.path(tempdir(), "df1.csv"))
# sm1_measuresB = stan_glmer(n_hl_measure ~ population_trend * n_hl_threat + (1|featurename),
# data = model_data_measure, diagnostic_file = file.path(tempdir(), "df2.csv"))
#
#
# library(bridgesampling)
# bridge_0 <- bridge_sampler(sm1_measures_NULL)
# bridge_1 <- bridge_sampler(sm1_measuresA)
# bridge_2 <- bridge_sampler(sm1_measuresB)
#
# bf(bridge_1, bridge_0)
# bf(bridge_2, bridge_1)
#
# plot(sm1_measuresA)In general, popualtions with more threats have significantly more measures, but this doesn’t appear to vary with the type of population trend (e.g. declines, increases, etc)
plot_model(m1_measuresA, show.intercept = TRUE)
tab_model(m1_measuresA) | n_hl_measure | |||
|---|---|---|---|
| Predictors | Estimates | CI | p |
| (Intercept) | 1.52 | 0.95 – 2.09 | <0.001 |
| population_trend_long [D] | -0.20 | -1.65 – 1.25 | 0.788 |
| population_trend_long [I] | -0.26 | -0.70 – 0.17 | 0.228 |
| population_trend_long [S] | -0.46 | -1.69 – 0.77 | 0.466 |
| population_trend_long [U] | -1.10 | -3.14 – 0.95 | 0.294 |
|
population_trend_long [Unk] |
-0.37 | -1.21 – 0.47 | 0.394 |
| n_hl_threat | 0.46 | 0.36 – 0.57 | <0.001 |
| Random Effects | |||
| σ2 | 2.10 | ||
| τ00 region | 0.25 | ||
| ICC | 0.10 | ||
| N region | 12 | ||
| Observations | 265 | ||
| Marginal R2 / Conditional R2 | 0.244 / 0.323 | ||
plot_model(m1_measuresA, type = "pred")## $population_trend_long
##
## $n_hl_threat
#plot_model(m1_measuresB, type = "pred", terms = c("mean_n_hl_threat", "population_trend")) # Hard to see significant effects on orediction plot because of broad CIs of D popsDeclines vs Threats/Measures
Are species more likely to be declining where there are more threats/measures?
# Model data is only declines or increases (remapped to 'increasing' as 0/1)
model_data_binom = model_data %>%
ungroup() %>%
mutate(population_trend = factor(population_trend)) %>%
filter(population_trend == "D" | population_trend == "I" | population_trend == "S") %>% # Subset to only increases or declines
mutate(increasing = ifelse(population_trend == "I", 1, 0)) %>% # Response variable with be 'increasing' or otherwise 'decreasing'
select(increasing, n_hl_threat, n_hl_pressure, n_hl_measure, featurename) %>% # Columns for modelling
drop_na() # Get rid of NAs in any of these columns
# Are populations more likely to be declining where there are more threats/measures
m1_NULL = glmer(increasing ~ (1|featurename),
data = model_data_binom, family = "binomial", control = glmerControl(optimizer = "bobyqa"))
# Are populations more likely to be declining where there are more threats/measures
m1_threat = glmer(increasing ~ scale(n_hl_threat) + (1|featurename),
data = model_data_binom, family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1_measure = glmer(increasing ~ scale(n_hl_measure) + (1|featurename),
data = model_data_binom, family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1_pressuee = glmer(increasing ~ scale(n_hl_pressure) + (1|featurename),
data = model_data_binom, family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1A = glmer(increasing ~ scale(n_hl_threat) + scale(n_hl_measure) + (1|featurename),
data = model_data_binom, family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1B = glmer(increasing ~ scale(n_hl_threat) * scale(n_hl_measure) + (1|featurename),
data = model_data_binom, family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1C = glmer(increasing ~ scale(n_hl_pressure) + scale(n_hl_measure) + (1|featurename),
data = model_data_binom, family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1D = glmer(increasing ~ scale(n_hl_pressure) * scale(n_hl_measure) + (1|featurename),
data = model_data_binom, family = "binomial", control = glmerControl(optimizer = "bobyqa"))
# Adding 'country' throws singularities due to sparse data
#summary(m1)
tab_model(m1_NULL, m1_threat, m1_measure, m1A, m1B, m1C, m1D, show.aicc = TRUE)| increasing | increasing | increasing | increasing | increasing | increasing | increasing | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Predictors | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p |
| (Intercept) | 1.49 | 0.63 – 3.53 | 0.368 | 1.51 | 0.64 – 3.59 | 0.348 | 1.46 | 0.61 – 3.46 | 0.393 | 1.51 | 0.64 – 3.57 | 0.346 | 1.51 | 0.63 – 3.63 | 0.358 | 1.50 | 0.64 – 3.50 | 0.351 | 1.49 | 0.63 – 3.55 | 0.368 |
| n_hl_threat | 1.12 | 0.82 – 1.55 | 0.473 | 1.43 | 0.96 – 2.13 | 0.077 | 1.43 | 0.95 – 2.17 | 0.087 | ||||||||||||
| n_hl_measure | 0.81 | 0.59 – 1.10 | 0.179 | 0.67 | 0.45 – 0.98 | 0.037 | 0.66 | 0.45 – 0.98 | 0.038 | 0.67 | 0.46 – 0.98 | 0.040 | 0.67 | 0.46 – 0.98 | 0.041 | ||||||
|
n_hl_threat * n_hl_measure |
1.00 | 0.73 – 1.37 | 0.981 | ||||||||||||||||||
| n_hl_pressure | 1.42 | 0.95 – 2.13 | 0.088 | 1.42 | 0.94 – 2.16 | 0.095 | |||||||||||||||
|
n_hl_pressure * n_hl_measure |
1.01 | 0.74 – 1.38 | 0.961 | ||||||||||||||||||
| Random Effects | |||||||||||||||||||||
| σ2 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | ||||||||||||||
| τ00 | 2.35 featurename | 2.34 featurename | 2.36 featurename | 2.30 featurename | 2.30 featurename | 2.24 featurename | 2.23 featurename | ||||||||||||||
| ICC | 0.42 | 0.42 | 0.42 | 0.41 | 0.41 | 0.40 | 0.40 | ||||||||||||||
| N | 18 featurename | 18 featurename | 18 featurename | 18 featurename | 18 featurename | 18 featurename | 18 featurename | ||||||||||||||
| Observations | 229 | 229 | 229 | 229 | 229 | 229 | 229 | ||||||||||||||
| Marginal R2 / Conditional R2 | 0.000 / 0.417 | 0.002 / 0.417 | 0.008 / 0.422 | 0.022 / 0.424 | 0.022 / 0.424 | 0.021 / 0.417 | 0.021 / 0.417 | ||||||||||||||
| AIC | 286.813 | 288.361 | 287.068 | 285.988 | 288.078 | 286.221 | 288.309 | ||||||||||||||
compare_performance(m1_NULL, m1_threat, m1_measure, m1A, m1B, m1C, m1D)## # Comparison of Model Performance Indices
##
## Name | Model | AIC | BIC | R2 (cond.) | R2 (marg.) | ICC | RMSE | Sigma | Log_loss | Score_log | Score_spherical
## ------------------------------------------------------------------------------------------------------------------------------------
## m1_NULL | glmerMod | 286.760 | 293.627 | 0.417 | 0.000 | 0.417 | 0.421 | 1.000 | 0.533 | -112.100 | 0.007
## m1_threat | glmerMod | 288.254 | 298.555 | 0.417 | 0.002 | 0.416 | 0.421 | 1.000 | 0.532 | -112.337 | 0.009
## m1_measure | glmerMod | 286.961 | 297.262 | 0.422 | 0.008 | 0.417 | 0.419 | 1.000 | 0.529 | -112.968 | 0.005
## m1A | glmerMod | 285.809 | 299.544 | 0.424 | 0.022 | 0.411 | 0.417 | 1.000 | 0.523 | -114.723 | 0.007
## m1B | glmerMod | 287.809 | 304.977 | 0.424 | 0.022 | 0.411 | 0.417 | 1.000 | 0.523 | -114.720 | 0.007
## m1C | glmerMod | 286.042 | 299.777 | 0.417 | 0.021 | 0.405 | 0.418 | 1.000 | 0.525 | -114.417 | 0.008
## m1D | glmerMod | 288.040 | 305.208 | 0.417 | 0.021 | 0.404 | 0.418 | 1.000 | 0.525 | -114.404 | 0.007Support for m1A (including both threats and measures, but no interaction)
plot_model(m1A, show.intercept = TRUE)
tab_model(m1A) | increasing | |||
|---|---|---|---|
| Predictors | Odds Ratios | CI | p |
| (Intercept) | 1.51 | 0.64 – 3.57 | 0.346 |
| n_hl_threat | 1.43 | 0.96 – 2.13 | 0.077 |
| n_hl_measure | 0.67 | 0.45 – 0.98 | 0.037 |
| Random Effects | |||
| σ2 | 3.29 | ||
| τ00 featurename | 2.30 | ||
| ICC | 0.41 | ||
| N featurename | 18 | ||
| Observations | 229 | ||
| Marginal R2 / Conditional R2 | 0.022 / 0.424 | ||
plot_model(m1A, type = "pred")## $n_hl_threat
##
## $n_hl_measure
Long-term change
Are species more likely to be declining where there are more threats/measures?
# Model data is only declines or increases (remapped to 'increasing' as 0/1)
model_data_binom_long = model_data %>%
ungroup() %>%
mutate(population_trend_long = factor(population_trend_long)) %>%
filter(population_trend_long == "D" | population_trend_long == "I") %>% # Subset to only increases or declines
mutate(increasing = ifelse(population_trend_long == "I", 1, 0)) %>% # Response variable with be 'increasing' or otherwise 'decreasing'
select(increasing, n_hl_threat, n_hl_measure, n_hl_pressure, featurename) %>% # Columns for modelling
drop_na() # Get rid of NAs in any of these columns
# Are populations more likely to be declining where there are more threats/measures
m2_NULL = glmer(increasing ~ (1|featurename),
data = model_data_binom_long, family = "binomial", control = glmerControl(optimizer = "bobyqa"))
# Are populations more likely to be declining where there are more threats/measures
m2_threat = glmer(increasing ~ scale(n_hl_threat) + (1|featurename),
data = model_data_binom_long, family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2_measure = glmer(increasing ~ scale(n_hl_measure) + (1|featurename),
data = model_data_binom_long, family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2_pressure = glmer(increasing ~ scale(n_hl_pressure) + (1|featurename),
data = model_data_binom_long, family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2A = glmer(increasing ~ scale(n_hl_threat) + scale(n_hl_measure) + (1|featurename),
data = model_data_binom_long, family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2B = glmer(increasing ~ scale(n_hl_threat) * scale(n_hl_measure) + (1|featurename),
data = model_data_binom_long, family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2C = glmer(increasing ~ scale(n_hl_threat) + scale(n_hl_pressure) + (1|featurename),
data = model_data_binom_long, family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2D = glmer(increasing ~ scale(n_hl_threat) * scale(n_hl_pressure) + (1|featurename),
data = model_data_binom_long, family = "binomial", control = glmerControl(optimizer = "bobyqa")) ## boundary (singular) fit: see ?isSingular# Adding 'country' throws singularities due to sparse data
#summary(m1)
tab_model(m2_NULL, m2_threat, m2_measure, m2A, m2B, m2C, m2D, show.aicc = TRUE)| increasing | increasing | increasing | increasing | increasing | increasing | increasing | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Predictors | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p |
| (Intercept) | 21.10 | 3.21 – 138.50 | 0.001 | 20.82 | 3.39 – 127.92 | 0.001 | 20.97 | 3.22 – 136.31 | 0.001 | 21.75 | 3.31 – 143.07 | 0.001 | 26.58 | 2.57 – 275.34 | 0.006 | 20.36 | 3.26 – 127.04 | 0.001 | 11.35 | 3.08 – 41.78 | <0.001 |
| n_hl_threat | 1.56 | 0.51 – 4.75 | 0.433 | 1.98 | 0.43 – 9.22 | 0.382 | 1.92 | 0.37 – 10.01 | 0.441 | 1.87 | 0.11 – 30.67 | 0.660 | 2.27 | 0.16 – 31.58 | 0.541 | ||||||
| n_hl_measure | 1.06 | 0.35 – 3.15 | 0.922 | 0.71 | 0.18 – 2.86 | 0.628 | 0.69 | 0.15 – 3.03 | 0.618 | ||||||||||||
|
n_hl_threat * n_hl_measure |
0.81 | 0.22 – 2.97 | 0.745 | ||||||||||||||||||
| n_hl_pressure | 0.82 | 0.05 – 12.90 | 0.889 | 1.11 | 0.10 – 12.12 | 0.930 | |||||||||||||||
|
n_hl_threat * n_hl_pressure |
2.15 | 0.38 – 12.10 | 0.384 | ||||||||||||||||||
| Random Effects | |||||||||||||||||||||
| σ2 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | ||||||||||||||
| τ00 | 0.75 featurename | 0.46 featurename | 0.73 featurename | 0.46 featurename | 0.66 featurename | 0.39 featurename | 0.00 featurename | ||||||||||||||
| ICC | 0.19 | 0.12 | 0.18 | 0.12 | 0.17 | 0.11 | |||||||||||||||
| N | 12 featurename | 12 featurename | 12 featurename | 12 featurename | 12 featurename | 12 featurename | 12 featurename | ||||||||||||||
| Observations | 68 | 68 | 68 | 68 | 68 | 68 | 68 | ||||||||||||||
| Marginal R2 / Conditional R2 | 0.000 / 0.186 | 0.050 / 0.166 | 0.001 / 0.182 | 0.078 / 0.192 | 0.090 / 0.241 | 0.053 / 0.153 | 0.274 / NA | ||||||||||||||
| AIC | 34.449 | 36.027 | 36.630 | 38.051 | 40.276 | 38.268 | 39.637 | ||||||||||||||
compare_performance(m2_NULL, m2_threat, m2_measure, m2A, m2B, m2C, m2D)## Random effect variances not available. Returned R2 does not account for random effects.## # Comparison of Model Performance Indices
##
## Name | Model | AIC | BIC | R2 (cond.) | R2 (marg.) | RMSE | Sigma | Log_loss | Score_log | Score_spherical | ICC
## ----------------------------------------------------------------------------------------------------------------------------------
## m2_NULL | glmerMod | 34.265 | 38.704 | 0.186 | 0.000 | 0.228 | 1.000 | 0.193 | -189.585 | 0.036 | 0.186
## m2_threat | glmerMod | 35.652 | 42.310 | 0.166 | 0.050 | 0.228 | 1.000 | 0.198 | -190.175 | 0.015 | 0.122
## m2_measure | glmerMod | 36.255 | 42.914 | 0.182 | 7.336e-04 | 0.228 | 1.000 | 0.194 | -189.298 | 0.017 | 0.181
## m2A | glmerMod | 37.416 | 46.294 | 0.192 | 0.078 | 0.229 | 1.000 | 0.197 | -192.934 | 0.015 | 0.123
## m2B | glmerMod | 39.308 | 50.406 | 0.241 | 0.090 | 0.227 | 1.000 | 0.190 | -197.320 | 0.015 | 0.166
## m2C | glmerMod | 37.633 | 46.511 | 0.153 | 0.053 | 0.229 | 1.000 | 0.201 | -189.137 | 0.015 | 0.106
## m2D | glmerMod | 38.669 | 49.767 | | 0.274 | 0.232 | 1.000 | 0.211 | -197.271 | 0.015 |No clear support for threats/measures helping predict long-term increases - null model preferred
plot_model(m2_NULL, show.intercept = TRUE)
tab_model(m2_NULL) | increasing | |||
|---|---|---|---|
| Predictors | Odds Ratios | CI | p |
| (Intercept) | 21.10 | 3.21 – 138.50 | 0.001 |
| Random Effects | |||
| σ2 | 3.29 | ||
| τ00 featurename | 0.75 | ||
| ICC | 0.19 | ||
| N featurename | 12 | ||
| Observations | 68 | ||
| Marginal R2 / Conditional R2 | 0.000 / 0.186 | ||
plot_model(m2_NULL, type = "pred")## list()Range-change vs Threats/Measures
Are species ranges more likely to be shrinking where there are more threats/measures?
model_data_binom_range = model_data %>%
ungroup() %>%
mutate(range_trend = factor(range_trend)) %>%
filter(range_trend == "D" | range_trend == "I" | range_trend == "S") %>% # Subset to only increases or declines
mutate(increasing = ifelse(range_trend == "I", 1, 0)) %>% # Response variable with be 'increasing' or otherwise 'decreasing'
select(region,increasing, n_hl_measure, n_hl_threat, n_hl_pressure, featurename) %>% # Columns for modelling
drop_na() # Get rid of NAs in any of these columns
# Are species ranges more likely to be shrinking where there are more threats/measures
m1r_NULL = glmer(increasing ~ (1|featurename),
data = model_data_binom_range,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1r_threat = glmer(increasing ~ scale(n_hl_threat) + (1|featurename),
data = model_data_binom_range,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1r_measure = glmer(increasing ~ scale(n_hl_measure) + (1|featurename),
data = model_data_binom_range,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1r_pressure = glmer(increasing ~ scale(n_hl_pressure) + (1|featurename),
data = model_data_binom_range,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1rA = glmer(increasing ~ scale(n_hl_measure) + scale(n_hl_threat) + (1|featurename),
data = model_data_binom_range,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1rB = glmer(increasing ~ scale(n_hl_measure) * scale(n_hl_threat) + (1|featurename),
data = model_data_binom_range,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1rC = glmer(increasing ~ scale(n_hl_measure) + scale(n_hl_pressure) + (1|featurename),
data = model_data_binom_range,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1rD = glmer(increasing ~ scale(n_hl_measure) * scale(n_hl_pressure) + (1|featurename),
data = model_data_binom_range,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
tab_model(m1r_NULL, m1r_threat, m1r_measure, m1rA, m1rB, m1rC, m1rD, show.aicc = TRUE)| increasing | increasing | increasing | increasing | increasing | increasing | increasing | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Predictors | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p |
| (Intercept) | 0.39 | 0.17 – 0.91 | 0.029 | 0.41 | 0.18 – 0.92 | 0.031 | 0.39 | 0.17 – 0.91 | 0.029 | 0.40 | 0.18 – 0.92 | 0.031 | 0.35 | 0.15 – 0.80 | 0.013 | 0.40 | 0.18 – 0.91 | 0.029 | 0.35 | 0.15 – 0.80 | 0.013 |
| n_hl_threat | 1.24 | 0.90 – 1.71 | 0.193 | 1.33 | 0.91 – 1.92 | 0.137 | 1.50 | 1.00 – 2.25 | 0.052 | ||||||||||||
| n_hl_measure | 1.00 | 0.74 – 1.37 | 0.976 | 0.88 | 0.61 – 1.25 | 0.470 | 0.81 | 0.56 – 1.18 | 0.278 | 0.88 | 0.61 – 1.26 | 0.472 | 0.82 | 0.56 – 1.19 | 0.300 | ||||||
|
n_hl_measure * n_hl_threat |
1.32 | 0.96 – 1.82 | 0.092 | ||||||||||||||||||
| n_hl_pressure | 1.32 | 0.91 – 1.92 | 0.146 | 1.47 | 0.98 – 2.21 | 0.064 | |||||||||||||||
|
n_hl_measure * n_hl_pressure |
1.28 | 0.94 – 1.76 | 0.121 | ||||||||||||||||||
| Random Effects | |||||||||||||||||||||
| σ2 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | ||||||||||||||
| τ00 | 2.35 featurename | 2.19 featurename | 2.35 featurename | 2.23 featurename | 2.15 featurename | 2.18 featurename | 2.09 featurename | ||||||||||||||
| ICC | 0.42 | 0.40 | 0.42 | 0.40 | 0.39 | 0.40 | 0.39 | ||||||||||||||
| N | 19 featurename | 19 featurename | 19 featurename | 19 featurename | 19 featurename | 19 featurename | 19 featurename | ||||||||||||||
| Observations | 253 | 253 | 253 | 253 | 253 | 253 | 253 | ||||||||||||||
| Marginal R2 / Conditional R2 | 0.000 / 0.417 | 0.008 / 0.404 | 0.000 / 0.417 | 0.010 / 0.410 | 0.024 / 0.409 | 0.010 / 0.405 | 0.022 / 0.403 | ||||||||||||||
| AIC | 280.443 | 280.833 | 282.491 | 282.387 | 281.639 | 282.488 | 282.181 | ||||||||||||||
compare_performance(m1r_NULL, m1r_threat, m1r_measure, m1rA, m1rB, m1rC, m1rD)## # Comparison of Model Performance Indices
##
## Name | Model | AIC | BIC | R2 (cond.) | R2 (marg.) | ICC | RMSE | Sigma | Log_loss | Score_log | Score_spherical
## -------------------------------------------------------------------------------------------------------------------------------------
## m1r_NULL | glmerMod | 280.395 | 287.462 | 0.417 | 0.000 | 0.417 | 0.393 | 1.000 | 0.468 | -48.904 | 0.023
## m1r_threat | glmerMod | 280.737 | 291.337 | 0.404 | 0.008 | 0.400 | 0.393 | 1.000 | 0.467 | -49.219 | 0.019
## m1r_measure | glmerMod | 282.395 | 292.995 | 0.417 | 3.996e-06 | 0.417 | 0.393 | 1.000 | 0.468 | -48.901 | 0.023
## m1rA | glmerMod | 282.226 | 296.360 | 0.410 | 0.010 | 0.404 | 0.392 | 1.000 | 0.466 | -49.406 | 0.019
## m1rB | glmerMod | 281.396 | 299.063 | 0.409 | 0.024 | 0.395 | 0.389 | 1.000 | 0.461 | -50.100 | 0.013
## m1rC | glmerMod | 282.327 | 296.460 | 0.405 | 0.010 | 0.399 | 0.392 | 1.000 | 0.467 | -49.335 | 0.020
## m1rD | glmerMod | 281.938 | 299.605 | 0.403 | 0.022 | 0.389 | 0.390 | 1.000 | 0.463 | -49.940 | 0.014No clear support for threats/measures helping predict range increases - null model preferred
plot_model(m1r_NULL, show.intercept = TRUE)
tab_model(m1r_NULL) | increasing | |||
|---|---|---|---|
| Predictors | Odds Ratios | CI | p |
| (Intercept) | 0.39 | 0.17 – 0.91 | 0.029 |
| Random Effects | |||
| σ2 | 3.29 | ||
| τ00 featurename | 2.35 | ||
| ICC | 0.42 | ||
| N featurename | 19 | ||
| Observations | 253 | ||
| Marginal R2 / Conditional R2 | 0.000 / 0.417 | ||
plot_model(m1r_NULL, type = "pred")## list()Are species long-term ranges more likely to be shrinking where there are more threats/measures?
model_data_binom_range_long = model_data %>%
ungroup() %>%
mutate(range_trend_long = factor(range_trend_long)) %>%
filter(range_trend_long == "D" | range_trend_long == "I" | range_trend_long == "S") %>% # Subset to only increases or declines
mutate(increasing = ifelse(range_trend_long == "I", 1, 0)) %>% # Response variable with be 'increasing' or otherwise 'decreasing'
select(region, increasing, n_hl_measure, n_hl_threat, n_hl_pressure, featurename) %>% # Columns for modelling
drop_na() # Get rid of NAs in any of these columns
# Are species ranges more likely to be shrinking where there are more threats/measures
m2r_NULL = glmer(increasing ~ (1|featurename),
data = model_data_binom_range_long,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2r_threat = glmer(increasing ~ scale(n_hl_threat) + (1|featurename),
data = model_data_binom_range_long,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2r_measure = glmer(increasing ~ scale(n_hl_measure) + (1|featurename),
data = model_data_binom_range_long,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2r_pressure = glmer(increasing ~ scale(n_hl_pressure) + (1|featurename),
data = model_data_binom_range_long,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2rA = glmer(increasing ~ scale(n_hl_measure) + scale(n_hl_threat) + (1|featurename),
data = model_data_binom_range_long,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2rB = glmer(increasing ~ scale(n_hl_measure) * scale(n_hl_threat) + (1|featurename),
data = model_data_binom_range_long,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2rC = glmer(increasing ~ scale(n_hl_measure) + scale(n_hl_pressure) + (1|featurename),
data = model_data_binom_range_long,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2rD = glmer(increasing ~ scale(n_hl_measure) * scale(n_hl_pressure) + (1|featurename),
data = model_data_binom_range_long,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
tab_model(m2r_NULL, m2r_threat, m2r_measure, m2r_pressure, m2rA, m2rB, m2rC, m2rD, show.aicc = TRUE)| increasing | increasing | increasing | increasing | increasing | increasing | increasing | increasing | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Predictors | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p |
| (Intercept) | 2.97 | 1.15 – 7.71 | 0.025 | 3.09 | 1.24 – 7.73 | 0.016 | 2.98 | 1.13 – 7.85 | 0.027 | 3.06 | 1.21 – 7.72 | 0.018 | 3.27 | 1.25 – 8.52 | 0.016 | 2.76 | 1.01 – 7.50 | 0.047 | 3.14 | 1.21 – 8.11 | 0.018 | 2.40 | 0.94 – 6.13 | 0.067 |
| n_hl_threat | 1.43 | 0.72 – 2.85 | 0.308 | 2.03 | 0.78 – 5.32 | 0.148 | 2.28 | 0.86 – 6.05 | 0.099 | |||||||||||||||
| n_hl_measure | 0.90 | 0.46 – 1.76 | 0.751 | 0.60 | 0.24 – 1.46 | 0.260 | 0.62 | 0.26 – 1.49 | 0.283 | 0.70 | 0.30 – 1.62 | 0.405 | 0.72 | 0.32 – 1.63 | 0.427 | |||||||||
| n_hl_pressure | 1.29 | 0.64 – 2.57 | 0.474 | 1.60 | 0.66 – 3.85 | 0.295 | 2.09 | 0.83 – 5.28 | 0.118 | |||||||||||||||
|
n_hl_measure * n_hl_threat |
1.41 | 0.60 – 3.30 | 0.433 | |||||||||||||||||||||
|
n_hl_measure * n_hl_pressure |
1.87 | 0.79 – 4.44 | 0.157 | |||||||||||||||||||||
| Random Effects | ||||||||||||||||||||||||
| σ2 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | ||||||||||||||||
| τ00 | 1.43 featurename | 1.13 featurename | 1.52 featurename | 1.21 featurename | 1.23 featurename | 1.08 featurename | 1.26 featurename | 0.97 featurename | ||||||||||||||||
| ICC | 0.30 | 0.26 | 0.32 | 0.27 | 0.27 | 0.25 | 0.28 | 0.23 | ||||||||||||||||
| N | 14 featurename | 14 featurename | 14 featurename | 14 featurename | 14 featurename | 14 featurename | 14 featurename | 14 featurename | ||||||||||||||||
| Observations | 68 | 68 | 68 | 68 | 68 | 68 | 68 | 68 | ||||||||||||||||
| Marginal R2 / Conditional R2 | 0.000 / 0.303 | 0.028 / 0.278 | 0.002 / 0.317 | 0.014 / 0.278 | 0.070 / 0.324 | 0.079 / 0.308 | 0.034 / 0.302 | 0.089 / 0.296 | ||||||||||||||||
| AIC | 75.374 | 76.554 | 77.468 | 77.070 | 77.485 | 79.193 | 78.620 | 78.831 | ||||||||||||||||
compare_performance(m2r_NULL, m2r_threat, m2r_measure, m2r_pressure, m2rA, m2rB, m2rC, m2rD)## # Comparison of Model Performance Indices
##
## Name | Model | AIC | BIC | R2 (cond.) | R2 (marg.) | ICC | RMSE | Sigma | Log_loss | Score_log | Score_spherical
## ------------------------------------------------------------------------------------------------------------------------------------
## m2r_NULL | glmerMod | 75.189 | 79.628 | 0.303 | 0.000 | 0.303 | 0.359 | 1.000 | 0.410 | -78.782 | 0.027
## m2r_threat | glmerMod | 76.179 | 82.838 | 0.278 | 0.028 | 0.257 | 0.361 | 1.000 | 0.418 | -79.701 | 0.021
## m2r_measure | glmerMod | 77.093 | 83.751 | 0.317 | 0.002 | 0.315 | 0.357 | 1.000 | 0.406 | -79.215 | 0.029
## m2r_pressure | glmerMod | 76.695 | 83.353 | 0.278 | 0.014 | 0.268 | 0.362 | 1.000 | 0.417 | -79.076 | 0.015
## m2rA | glmerMod | 76.850 | 85.728 | 0.324 | 0.070 | 0.273 | 0.352 | 1.000 | 0.404 | -83.118 | 0.015
## m2rB | glmerMod | 78.225 | 89.322 | 0.308 | 0.079 | 0.248 | 0.356 | 1.000 | 0.408 | -83.012 | 0.015
## m2rC | glmerMod | 77.986 | 86.864 | 0.302 | 0.034 | 0.278 | 0.358 | 1.000 | 0.410 | -80.732 | 0.016
## m2rD | glmerMod | 77.864 | 88.961 | 0.296 | 0.089 | 0.228 | 0.361 | 1.000 | 0.411 | -82.678 | 0.015No clear support for threats/measures helping predict long-term range increases - null model preferred
plot_model(m2r_NULL, show.intercept = TRUE)
tab_model(m2r_NULL) | increasing | |||
|---|---|---|---|
| Predictors | Odds Ratios | CI | p |
| (Intercept) | 2.97 | 1.15 – 7.71 | 0.025 |
| Random Effects | |||
| σ2 | 3.29 | ||
| τ00 featurename | 1.43 | ||
| ICC | 0.30 | ||
| N featurename | 14 | ||
| Observations | 68 | ||
| Marginal R2 / Conditional R2 | 0.000 / 0.303 | ||
plot_model(m2r_NULL, type = "pred")## list()Habitat-change vs Threats/Measures
Are species habitats more likely to be shrinking where there are more threats/measures?
model_data_binom_habitat = model_data %>%
ungroup() %>%
mutate(habitat_trend = factor(habitat_trend)) %>%
filter(habitat_trend == "D" | habitat_trend == "I") %>% # Subset to only increases or declines
mutate(increasing = ifelse(habitat_trend == "I", 1, 0)) %>% # Response variable with be 'increasing' or otherwise 'decreasing'
select(region, increasing, n_hl_measure, n_hl_threat, n_hl_pressure, featurename) %>% # Columns for modelling
drop_na() # Get rid of NAs in any of these columns
# Are species ranges more likely to be shrinking where there are more threats/measures
m1h_NULL = glmer(increasing ~ (1|featurename),
data = model_data_binom_habitat,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1h_threat = glmer(increasing ~ scale(n_hl_threat) + (1|featurename),
data = model_data_binom_habitat,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1h_measure = glmer(increasing ~ scale(n_hl_measure) + (1|featurename),
data = model_data_binom_habitat,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1h_pressure = glmer(increasing ~ scale(n_hl_pressure) + (1|featurename),
data = model_data_binom_habitat,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1hA = glmer(increasing ~ scale(n_hl_measure) + scale(n_hl_threat) + (1|featurename),
data = model_data_binom_habitat,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1hB = glmer(increasing ~ scale(n_hl_measure) * scale(n_hl_threat) + (1|featurename),
data = model_data_binom_habitat,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1hC = glmer(increasing ~ scale(n_hl_measure) + scale(n_hl_pressure) + (1|featurename),
data = model_data_binom_habitat,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m1hD = glmer(increasing ~ scale(n_hl_measure) * scale(n_hl_pressure) + (1|featurename),
data = model_data_binom_habitat,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
tab_model(m1h_NULL, m1h_threat, m1h_measure, m1h_pressure, m1hA, m1hB, m1hC, m1hD, show.aicc = TRUE)| increasing | increasing | increasing | increasing | increasing | increasing | increasing | increasing | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Predictors | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p |
| (Intercept) | 18.57 | 0.35 – 979.19 | 0.149 | 84.89 | 0.02 – 299367.00 | 0.287 | 257.20 | 0.10 – 682828.69 | 0.168 | 34.81 | 0.18 – 6793.30 | 0.187 | 291.09 | 0.06 – 1522604.31 | 0.194 | 1099.33 | 0.15 – 7808897.08 | 0.122 | 1359.38 | 0.01 – 246625716.90 | 0.243 | 461.71 | 0.04 – 5261651.27 | 0.198 |
| n_hl_threat | 0.25 | 0.01 – 5.36 | 0.377 | 0.86 | 0.02 – 47.92 | 0.941 | 0.12 | 0.00 – 67.64 | 0.510 | |||||||||||||||
| n_hl_measure | 0.08 | 0.00 – 4.10 | 0.209 | 0.08 | 0.00 – 5.71 | 0.250 | 0.34 | 0.01 – 21.46 | 0.607 | 0.02 | 0.00 – 78.35 | 0.340 | 0.04 | 0.00 – 24.49 | 0.331 | |||||||||
| n_hl_pressure | 0.41 | 0.06 – 2.65 | 0.350 | 3.48 | 0.03 – 397.31 | 0.606 | 2.85 | 0.03 – 244.72 | 0.645 | |||||||||||||||
|
n_hl_measure * n_hl_threat |
0.05 | 0.00 – 4.91 | 0.199 | |||||||||||||||||||||
|
n_hl_measure * n_hl_pressure |
0.45 | 0.03 – 6.17 | 0.553 | |||||||||||||||||||||
| Random Effects | ||||||||||||||||||||||||
| σ2 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | ||||||||||||||||
| τ00 | 10.54 featurename | 26.53 featurename | 42.55 featurename | 16.50 featurename | 45.35 featurename | 45.60 featurename | 69.16 featurename | 42.07 featurename | ||||||||||||||||
| ICC | 0.76 | 0.89 | 0.93 | 0.83 | 0.93 | 0.93 | 0.95 | 0.93 | ||||||||||||||||
| N | 10 featurename | 10 featurename | 10 featurename | 10 featurename | 10 featurename | 10 featurename | 10 featurename | 10 featurename | ||||||||||||||||
| Observations | 44 | 44 | 44 | 44 | 44 | 44 | 44 | 44 | ||||||||||||||||
| Marginal R2 / Conditional R2 | 0.000 / 0.762 | 0.060 / 0.896 | 0.121 / 0.937 | 0.038 / 0.840 | 0.121 / 0.941 | 0.176 / 0.945 | 0.144 / 0.961 | 0.150 / 0.938 | ||||||||||||||||
| AIC | 35.164 | 35.806 | 32.855 | 36.354 | 35.272 | 34.510 | 34.936 | 37.202 | ||||||||||||||||
compare_performance(m1h_NULL, m1h_threat, m1h_measure, m1h_pressure, m1hA, m1hB, m1hC, m1hD)## # Comparison of Model Performance Indices
##
## Name | Model | AIC | BIC | R2 (cond.) | R2 (marg.) | ICC | RMSE | Sigma | Log_loss | Score_log | Score_spherical
## ------------------------------------------------------------------------------------------------------------------------------------
## m1h_NULL | glmerMod | 34.872 | 38.440 | 0.762 | 0.000 | 0.762 | 0.225 | 1.000 | 0.179 | -69.197 | 0.072
## m1h_threat | glmerMod | 35.206 | 40.559 | 0.896 | 0.060 | 0.890 | 0.198 | 1.000 | 0.133 | -88.502 | 0.061
## m1h_measure | glmerMod | 32.255 | 37.608 | 0.937 | 0.121 | 0.928 | 0.175 | 1.000 | 0.098 | -118.506 | 0.061
## m1h_pressure | glmerMod | 35.754 | 41.107 | 0.840 | 0.038 | 0.834 | 0.214 | 1.000 | 0.152 | -77.985 | 0.062
## m1hA | glmerMod | 34.247 | 41.383 | 0.941 | 0.121 | 0.932 | 0.173 | 1.000 | 0.097 | -120.294 | 0.061
## m1hB | glmerMod | 32.931 | 41.852 | 0.945 | 0.176 | 0.933 | 0.157 | 1.000 | 0.079 | -136.775 | 0.065
## m1hC | glmerMod | 33.911 | 41.047 | 0.961 | 0.144 | 0.955 | 0.167 | 1.000 | 0.088 | -148.091 | 0.064
## m1hD | glmerMod | 35.623 | 44.544 | 0.938 | 0.150 | 0.927 | 0.176 | 1.000 | 0.098 | -125.941 | 0.062Some (weak) evidence that populations with more measures in place have declining habitat trends.
plot_model(m1h_measure, show.intercept = TRUE)
tab_model(m1h_measure) | increasing | |||
|---|---|---|---|
| Predictors | Odds Ratios | CI | p |
| (Intercept) | 257.20 | 0.10 – 682828.69 | 0.168 |
| n_hl_measure | 0.08 | 0.00 – 4.10 | 0.209 |
| Random Effects | |||
| σ2 | 3.29 | ||
| τ00 featurename | 42.55 | ||
| ICC | 0.93 | ||
| N featurename | 10 | ||
| Observations | 44 | ||
| Marginal R2 / Conditional R2 | 0.121 / 0.937 | ||
plot_model(m1h_measure, type = "pred")## $n_hl_measure
Are species long-term habitats more likely to be shrinking where there are more threats/measures?
model_data_binom_habitat_long = model_data %>%
ungroup() %>%
mutate(habitat_trend_long = factor(habitat_trend_long)) %>%
filter(habitat_trend_long == "D" | habitat_trend_long == "I" | habitat_trend_long == "S") %>% # Subset to only increases or declines
mutate(increasing = ifelse(habitat_trend_long == "I", 1, 0)) %>% # Response variable with be 'increasing' or otherwise 'decreasing'
select(region, increasing, n_hl_measure, n_hl_threat, n_hl_pressure, featurename) %>% # Columns for modelling
drop_na() # Get rid of NAs in any of these columns
# Are species ranges more likely to be shrinking where there are more threats/measures
m2h_NULL = glmer(increasing ~ (1|featurename),
data = model_data_binom_habitat_long,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2h_threat = glmer(increasing ~ scale(n_hl_threat) + (1|featurename),
data = model_data_binom_habitat_long,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2h_measure = glmer(increasing ~ scale(n_hl_measure) + (1|featurename),
data = model_data_binom_habitat_long,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2h_pressure = glmer(increasing ~ scale(n_hl_pressure) + (1|featurename),
data = model_data_binom_habitat_long,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2hA = glmer(increasing ~ scale(n_hl_measure) + scale(n_hl_threat) + (1|featurename),
data = model_data_binom_habitat_long,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2hB = glmer(increasing ~ scale(n_hl_measure) * scale(n_hl_threat) + (1|featurename),
data = model_data_binom_habitat_long,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2hC = glmer(increasing ~ scale(n_hl_measure) + scale(n_hl_pressure) + (1|featurename),
data = model_data_binom_habitat_long,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
m2hD = glmer(increasing ~ scale(n_hl_measure) * scale(n_hl_pressure) + (1|featurename),
data = model_data_binom_habitat_long,
family = "binomial", control = glmerControl(optimizer = "bobyqa"))
tab_model(m2h_NULL, m2h_threat, m2h_measure, m2h_pressure, m2hA, m2hB, m2hC, m2hD, show.aicc = TRUE)| increasing | increasing | increasing | increasing | increasing | increasing | increasing | increasing | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Predictors | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p | Odds Ratios | CI | p |
| (Intercept) | 0.46 | 0.20 – 1.05 | 0.065 | 0.45 | 0.19 – 1.05 | 0.063 | 0.47 | 0.21 – 1.03 | 0.060 | 0.46 | 0.20 – 1.04 | 0.062 | 0.45 | 0.19 – 1.05 | 0.065 | 0.49 | 0.20 – 1.21 | 0.124 | 0.46 | 0.20 – 1.04 | 0.061 | 0.53 | 0.22 – 1.26 | 0.152 |
| n_hl_threat | 1.48 | 0.75 – 2.93 | 0.263 | 1.50 | 0.65 – 3.48 | 0.343 | 1.41 | 0.60 – 3.35 | 0.430 | |||||||||||||||
| n_hl_measure | 1.22 | 0.66 – 2.25 | 0.534 | 0.98 | 0.45 – 2.11 | 0.951 | 1.04 | 0.47 – 2.29 | 0.925 | 1.05 | 0.51 – 2.16 | 0.885 | 1.20 | 0.56 – 2.56 | 0.636 | |||||||||
| n_hl_pressure | 1.38 | 0.72 – 2.66 | 0.335 | 1.34 | 0.63 – 2.86 | 0.445 | 1.20 | 0.54 – 2.68 | 0.657 | |||||||||||||||
|
n_hl_measure * n_hl_threat |
0.82 | 0.41 – 1.67 | 0.594 | |||||||||||||||||||||
|
n_hl_measure * n_hl_pressure |
0.68 | 0.33 – 1.38 | 0.288 | |||||||||||||||||||||
| Random Effects | ||||||||||||||||||||||||
| σ2 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | 3.29 | ||||||||||||||||
| τ00 | 0.43 featurename | 0.46 featurename | 0.35 featurename | 0.40 featurename | 0.47 featurename | 0.44 featurename | 0.39 featurename | 0.42 featurename | ||||||||||||||||
| ICC | 0.12 | 0.12 | 0.10 | 0.11 | 0.13 | 0.12 | 0.11 | 0.11 | ||||||||||||||||
| N | 11 featurename | 11 featurename | 11 featurename | 11 featurename | 11 featurename | 11 featurename | 11 featurename | 11 featurename | ||||||||||||||||
| Observations | 50 | 50 | 50 | 50 | 50 | 50 | 50 | 50 | ||||||||||||||||
| Marginal R2 / Conditional R2 | 0.000 / 0.116 | 0.039 / 0.158 | 0.010 / 0.107 | 0.027 / 0.134 | 0.039 / 0.160 | 0.052 / 0.165 | 0.028 / 0.130 | 0.068 / 0.173 | ||||||||||||||||
| AIC | 67.774 | 68.707 | 69.656 | 69.078 | 71.070 | 73.259 | 71.424 | 72.705 | ||||||||||||||||
compare_performance(m2h_NULL, m2h_threat, m2h_measure, m2h_pressure, m2hA, m2hB, m2hC, m2hD)## # Comparison of Model Performance Indices
##
## Name | Model | AIC | BIC | R2 (cond.) | R2 (marg.) | ICC | RMSE | Sigma | Log_loss | Score_log | Score_spherical
## ------------------------------------------------------------------------------------------------------------------------------------
## m2h_NULL | glmerMod | 67.518 | 71.343 | 0.116 | 0.000 | 0.116 | 0.441 | 1.000 | 0.571 | -6.203 | 0.087
## m2h_threat | glmerMod | 68.185 | 73.921 | 0.158 | 0.039 | 0.123 | 0.433 | 1.000 | 0.555 | -6.316 | 0.061
## m2h_measure | glmerMod | 69.134 | 74.870 | 0.107 | 0.010 | 0.097 | 0.443 | 1.000 | 0.576 | -6.250 | 0.085
## m2h_pressure | glmerMod | 68.556 | 74.292 | 0.134 | 0.027 | 0.109 | 0.437 | 1.000 | 0.565 | -6.288 | 0.068
## m2hA | glmerMod | 70.182 | 77.830 | 0.160 | 0.039 | 0.126 | 0.432 | 1.000 | 0.554 | -6.315 | 0.061
## m2hB | glmerMod | 71.895 | 81.456 | 0.165 | 0.052 | 0.119 | 0.432 | 1.000 | 0.554 | -6.330 | 0.052
## m2hC | glmerMod | 70.535 | 78.183 | 0.130 | 0.028 | 0.105 | 0.438 | 1.000 | 0.567 | -6.293 | 0.069
## m2hD | glmerMod | 71.341 | 80.901 | 0.173 | 0.068 | 0.113 | 0.430 | 1.000 | 0.552 | -6.380 | 0.054This doesn’t appear to be the case for the long-term trends, no sig. relationship between threats/measures and habitat change
plot_model(m2h_NULL, show.intercept = TRUE)
tab_model(m2h_NULL) | increasing | |||
|---|---|---|---|
| Predictors | Odds Ratios | CI | p |
| (Intercept) | 0.46 | 0.20 – 1.05 | 0.065 |
| Random Effects | |||
| σ2 | 3.29 | ||
| τ00 featurename | 0.43 | ||
| ICC | 0.12 | ||
| N featurename | 11 | ||
| Observations | 50 | ||
| Marginal R2 / Conditional R2 | 0.000 / 0.116 | ||
plot_model(m2h_NULL, type = "pred")## list()Notes
- Missing species? (these aren’t in Article 12 I think )
- Do we have enough information to weight these species by pop size?
sessionInfo()## R version 4.0.3 (2020-10-10)
## Platform: x86_64-apple-darwin17.0 (64-bit)
## Running under: macOS Big Sur 10.16
##
## Matrix products: default
## BLAS: /Library/Frameworks/R.framework/Versions/4.0/Resources/lib/libRblas.dylib
## LAPACK: /Library/Frameworks/R.framework/Versions/4.0/Resources/lib/libRlapack.dylib
##
## locale:
## [1] en_GB.UTF-8/en_GB.UTF-8/en_GB.UTF-8/C/en_GB.UTF-8/en_GB.UTF-8
##
## attached base packages:
## [1] stats graphics grDevices utils datasets methods base
##
## other attached packages:
## [1] performance_0.7.0 tidyr_1.1.2 sjPlot_2.8.7 lme4_1.1-26
## [5] Matrix_1.2-18 ggplot2_3.3.3 data.table_1.13.6 expss_0.10.7
## [9] dplyr_1.0.7 plyr_1.8.6
##
## loaded via a namespace (and not attached):
## [1] splines_4.0.3 carData_3.0-4 modelr_0.1.8 assertthat_0.2.1
## [5] statmod_1.4.35 highr_0.9 cellranger_1.1.0 yaml_2.2.1
## [9] bayestestR_0.8.2 pillar_1.6.1 backports_1.2.1 lattice_0.20-41
## [13] glue_1.4.2 digest_0.6.27 RColorBrewer_1.1-2 ggsignif_0.6.1
## [17] checkmate_2.0.0 minqa_1.2.4 colorspace_2.0-0 htmltools_0.5.1.1
## [21] pkgconfig_2.0.3 broom_0.7.5 haven_2.3.1 purrr_0.3.4
## [25] xtable_1.8-4 mvtnorm_1.1-1 scales_1.1.1 openxlsx_4.2.3
## [29] rio_0.5.26 emmeans_1.5.4 htmlTable_2.1.0 tibble_3.1.2
## [33] farver_2.1.0 generics_0.1.0 car_3.0-10 sjlabelled_1.1.7
## [37] ellipsis_0.3.2 ggpubr_0.4.0 withr_2.4.2 cli_3.0.0
## [41] readxl_1.3.1 magrittr_2.0.1 crayon_1.4.1 effectsize_0.4.3
## [45] estimability_1.3 evaluate_0.14 fansi_0.5.0 nlme_3.1-149
## [49] MASS_7.3-53 rstatix_0.7.0 forcats_0.5.1 foreign_0.8-81
## [53] tools_4.0.3 hms_1.0.0 lifecycle_1.0.0 matrixStats_0.58.0
## [57] stringr_1.4.0 munsell_0.5.0 zip_2.2.0 ggeffects_1.0.1
## [61] compiler_4.0.3 rlang_0.4.11 grid_4.0.3 nloptr_1.2.2.2
## [65] parameters_0.11.0 rstudioapi_0.13 htmlwidgets_1.5.3 labeling_0.4.2
## [69] rmarkdown_2.6 boot_1.3-25 gtable_0.3.0 abind_1.4-5
## [73] sjstats_0.18.1 DBI_1.1.1 curl_4.3.2 sjmisc_2.8.6
## [77] R6_2.5.0 knitr_1.33 utf8_1.2.1 insight_0.12.0
## [81] stringi_1.6.2 Rcpp_1.0.7 vctrs_0.3.8 tidyselect_1.1.1
## [85] xfun_0.24 coda_0.19-4