In karinorman/biodivTS: What the Package Does (One Line, Title Case)

Sensitivity Analysis

We performed sensitivity analyses to assess whether or not the general trends for biodiversity metrics changed based on major data processing decisions. We looked at 1) the minimum duration of included time series, 2) the minimum cut off for percentage of species with traits from a time series, and 3) in inclusion of amphibian data. The table below summarizes all models run and the variables for which a significant trend was detected. A star indicates a trend that was not detected in the original analyses presented in the manuscript. The model_id corresponds to the model_id in the CSV of sensitivity model results.

library(dplyr)
library(kableExtra)

tibble(sensitivity_test = c(rep("duration", 4), rep("trait coverage", 3)), 
       model_id = c("min3", "min4", "min5", "min10", "covg75", "covg85", "covg90"),
       description = c("Exclude time series with fewer than 3 years",
                       "Exclude time series with fewer than 4 years",
                       "Exclude time series with fewer than 5 years",
                       "Exclude time series with fewer than 10 years",
                       "Exclude time series with less than 75% of species with trait data",
                       "Exclude time series with less than 85% of species with trait data",
                       "Exclude time series with less than 90% of species with trait data"),
       sig_metrics = c("CWM body mass*, CWM fish diet, Jaccard", 
       "CWM body mass*, CWM fish diet, FRic*, Jaccard", 
       "CWM body mass*, CWM fish diet, FRic*, Jaccard",
       "CWM body mass*, CWM fish diet, Jaccard", rep("Jaccard, CWM fish diet", 3))
) %>% 
  #kable(format = "latex", caption = "Summary of the results for the sensitivity analyses, including those for duration of time series, trait coverage, and amphibian inclusion.") %>%
  kable(format = "latex") %>%
  column_spec(3:4, width = "3cm") %>%
  collapse_rows(columns = 1, valign = "middle") %>%
  kable_paper() %>%
  kable_styling(latex_options= c("hold_position")) #%>%
  #save_kable(here::here("figures/tables/sensitivity_table.jpeg"))

Amphibian Trends

There were 3 amphibian studies comprised of 11 time series in the bioTIME database with sufficient trait coverage to be included in the analysis. With such limited data, we excluded amphibians from the main results, but present preliminary trends here. Models fit with amphibian, mammal, and bird data combined showed qualitatively similar trend results to those reported in the manuscript that did not include amphibians. For models fit with only amphibian data, no general trends were detected for any metric. Table 2 gives model estimates for species-based and species corrected metrics.

readr::read_csv(here::here("figures/amph_model_table.csv")) %>%
  select(metric, everything(), -REML, -bobyqa) %>%
  select(-c(n_ts, formula, model)) %>%
  filter(group != "Residual" | is.na(group), metric %in% c("S", "Jaccard_base", "SES_FRic", "SES_FEve", "SES_FDiv")) %>%
  mutate(metric = ifelse(metric == "Jaccard_base", "log(Jaccard + 1)", metric),
         metric = ifelse(metric == "S", "log(Species Richness)", metric),
         effect = ifelse(effect == "ran_pars", "random", effect),
         group = case_when(
           group == "rarefyID:study_id" ~ "time series within study",
           group == "study_id" ~ "study",
           group == "Residual" ~ "residual",
           TRUE ~ group),
         term = case_when(
           term == "(Intercept)" ~ "Intercept",
           term == "year_scaled" ~ "Year",
           term == "sd__(Intercept)" ~ "SD Intercept",
           term == "cor__(Intercept).year_scaled" ~ "Corr(Intercept, Year)",
           term == "sd__year_scaled" ~ "SD Year",
           term == "sd__Observation" ~ "SD Observation",
           TRUE ~ term
         )) %>%
  mutate(across(c(estimate, std.error, p.value), round, 2),
         p.value = ifelse(p.value == 0, "<0.001", p.value)) %>%
  arrange(metric != "Species Richness", metric, !is.na(group), group, term == "Corr(Intercept, Year)", term) %>%
  dplyr::rename(grouping = group) %>%
  #kable(format = "latex", caption = "Model estimates and statistics for general trend models for species richness, Jaccard similarity, and standardized functional diversity metrics for Amphibian time series.") %>%
  kable(format = "latex") %>%
  column_spec(3, width = "2cm") %>%
  collapse_rows(columns = 1:3, valign = "middle") %>%
  kable_paper() %>%
  kable_styling(latex_options= c("scale_down", "hold_position")) #%>%
  #save_kable(here::here("figures/tables/amph_table.jpeg"))

Missing Individual Study Models

We fit models separately for each study and metric to estimate study-level trends. 11 models could not be fit because there were to many time series relative to the number of observations, and therefore not enough data to fit the appropriate mixed effects model structure. The missing models are in Table 3.

readr::read_csv(here::here("paper/missing_indv_studies.csv")) %>%
  rename(`n time series` = nrarefyID, `n observations` = n) %>%
  #kable(format = "latex", caption = "Model estimates and statistics for general trend models for species richness, Jaccard similarity, and standardized functional diversity metrics for Amphibian time series.") %>%
  kable(format = "latex") %>%
  kable_paper() %>%
  kable_styling(latex_options= c("scale_down", "hold_position"))# %>%
  #save_kable(here::here("figures/tables/missing_models_table.jpeg"))