Column inventories

workflowr

• Summary
• Checks
• Past versions

Last updated: 2022-11-14

Checks: 7 0

Knit directory: emlr_obs_analysis/analysis/

This reproducible R Markdown analysis was created with workflowr (version 1.7.0). The Checks tab describes the reproducibility checks that were applied when the results were created. The Past versions tab lists the development history.

---

R Markdown file: up-to-date

Great! Since the R Markdown file has been committed to the Git repository, you know the exact version of the code that produced these results.

Environment: empty

Great job! The global environment was empty. Objects defined in the global environment can affect the analysis in your R Markdown file in unknown ways. For reproduciblity it’s best to always run the code in an empty environment.

Seed: set.seed(20210412)

The command set.seed(20210412) was run prior to running the code in the R Markdown file. Setting a seed ensures that any results that rely on randomness, e.g. subsampling or permutations, are reproducible.

Session information: recorded

Great job! Recording the operating system, R version, and package versions is critical for reproducibility.

Cache: none

Nice! There were no cached chunks for this analysis, so you can be confident that you successfully produced the results during this run.

File paths: relative

Great job! Using relative paths to the files within your workflowr project makes it easier to run your code on other machines.

Repository version: 40f357d

Great! You are using Git for version control. Tracking code development and connecting the code version to the results is critical for reproducibility.

The results in this page were generated with repository version 40f357d. See the Past versions tab to see a history of the changes made to the R Markdown and HTML files.

Note that you need to be careful to ensure that all relevant files for the analysis have been committed to Git prior to generating the results (you can use wflow_publish or wflow_git_commit). workflowr only checks the R Markdown file, but you know if there are other scripts or data files that it depends on. Below is the status of the Git repository when the results were generated:

Ignored files:
    Ignored:    .Rhistory
    Ignored:    .Rproj.user/
    Ignored:    data/
    Ignored:    output/other/
    Ignored:    output/presentation/
    Ignored:    output/publication/

Untracked files:
    Untracked:  code/results_publication_backup_incl_ensemble_uncertainty_20221111.Rmd

Unstaged changes:
    Deleted:    analysis/MLR_target_budgets.Rmd
    Deleted:    analysis/MLR_target_column_inventories.Rmd
    Deleted:    analysis/MLR_target_zonal_sections.Rmd
    Modified:   analysis/_site.yml
    Modified:   code/Workflowr_project_managment.R

Note that any generated files, e.g. HTML, png, CSS, etc., are not included in this status report because it is ok for generated content to have uncommitted changes.

---

These are the previous versions of the repository in which changes were made to the R Markdown (analysis/WOA18_column_inventories.Rmd) and HTML (docs/WOA18_column_inventories.html) files. If you’ve configured a remote Git repository (see ?wflow_git_remote), click on the hyperlinks in the table below to view the files as they were in that past version.

File	Version	Author	Date	Message
html	cc337dd	jens-daniel-mueller	2022-11-11	Build site.
Rmd	c53fea0	jens-daniel-mueller	2022-11-11	rebuild website with woa18 clim

---

version_id_pattern <- "n"
config <- "MLR_basins"

1 Read files

print(version_id_pattern)

[1] "n"

# identify required version IDs

Version_IDs_1 <- list.files(path = "/nfs/kryo/work/jenmueller/emlr_cant/observations",
                            pattern = paste0("v_1", "n"))

Version_IDs_2 <- list.files(path = "/nfs/kryo/work/jenmueller/emlr_cant/observations",
                            pattern = paste0("v_2", "n"))

Version_IDs_3 <- list.files(path = "/nfs/kryo/work/jenmueller/emlr_cant/observations",
                            pattern = paste0("v_3", "n"))

Version_IDs <- c(Version_IDs_1, Version_IDs_2, Version_IDs_3)

print(Version_IDs)

 [1] "v_1n01" "v_1n02" "v_1n03" "v_1n04" "v_1n05" "v_1n06" "v_2n01" "v_2n02"
 [9] "v_2n03" "v_2n04" "v_2n05" "v_2n06" "v_3n01" "v_3n02" "v_3n03" "v_3n04"
[17] "v_3n05" "v_3n06"

for (i_Version_IDs in Version_IDs) {
  # i_Version_IDs <- Version_IDs[1]
  
  # print(i_Version_IDs)
  
  path_version_data     <-
    paste(path_observations,
          i_Version_IDs,
          "/data/",
          sep = "")
  
  # load and join data files
  
  dcant_inv <-
    read_csv(paste(path_version_data,
                   "dcant_inv.csv",
                   sep = ""))
  
  dcant_inv_mod_truth <-
    read_csv(paste(path_version_data,
                   "dcant_inv_mod_truth.csv",
                   sep = "")) %>%
    filter(method == "total") %>%
    select(-method)
  
  dcant_inv_bias <-
    read_csv(paste(path_version_data,
                   "dcant_inv_bias.csv",
                   sep = "")) %>%
    mutate(Version_ID = i_Version_IDs)
  
  dcant_inv <- bind_rows(dcant_inv,
                         dcant_inv_mod_truth) %>%
    mutate(Version_ID = i_Version_IDs)
  
  dcant_budget_lat_grid <-
    read_csv(paste(path_version_data,
                   "dcant_budget_lat_grid.csv",
                   sep = "")) %>%
    mutate(Version_ID = i_Version_IDs)
  
  dcant_budget_lon_grid <-
    read_csv(paste(path_version_data,
                   "dcant_budget_lon_grid.csv",
                   sep = "")) %>%
    mutate(Version_ID = i_Version_IDs)
  
  
  params_local <-
    read_rds(paste(path_version_data,
                   "params_local.rds",
                   sep = ""))
  
  params_local <- bind_cols(
    Version_ID = i_Version_IDs,
    MLR_basins := str_c(params_local$MLR_basins, collapse = "|"),
    tref1 = params_local$tref1,
    tref2 = params_local$tref2)
  
  tref <- read_csv(paste(path_version_data,
                         "tref.csv",
                         sep = ""))
  
  params_local <- params_local %>%
    mutate(
      median_year_1 = sort(tref$median_year)[1],
      median_year_2 = sort(tref$median_year)[2],
      duration = median_year_2 - median_year_1,
      period = paste(median_year_1, "-", median_year_2)
    )
  
  if (exists("dcant_inv_all")) {
    dcant_inv_all <- bind_rows(dcant_inv_all, dcant_inv)
  }
  
  if (!exists("dcant_inv_all")) {
    dcant_inv_all <- dcant_inv
  }
  
  if (exists("dcant_inv_bias_all")) {
    dcant_inv_bias_all <- bind_rows(dcant_inv_bias_all, dcant_inv_bias)
  }
  
  if (!exists("dcant_inv_bias_all")) {
    dcant_inv_bias_all <- dcant_inv_bias
  }
  
  if (exists("dcant_budget_lat_grid_all")) {
    dcant_budget_lat_grid_all <- bind_rows(dcant_budget_lat_grid_all, dcant_budget_lat_grid)
  }
  
  if (!exists("dcant_budget_lat_grid_all")) {
    dcant_budget_lat_grid_all <- dcant_budget_lat_grid
  }
  
  if (exists("dcant_budget_lon_grid_all")) {
    dcant_budget_lon_grid_all <- bind_rows(dcant_budget_lon_grid_all, dcant_budget_lon_grid)
  }
  
  if (!exists("dcant_budget_lon_grid_all")) {
    dcant_budget_lon_grid_all <- dcant_budget_lon_grid
  }

  if (exists("params_local_all")) {
    params_local_all <- bind_rows(params_local_all, params_local)
  }
  
  if (!exists("params_local_all")) {
    params_local_all <- params_local
  }
  
  
}

rm(dcant_inv,
   dcant_inv_bias,
   dcant_inv_mod_truth,
   dcant_budget_lat_grid,
   dcant_budget_lon_grid,
   params_local,
   tref)

# params_local_all <-
#   params_local_all %>%
#   mutate(period = factor(period, c("1994 - 2004", "2004 - 2014", "1994 - 2014")))

dcant_inv_all <- dcant_inv_all %>%
  filter(inv_depth == params_global$inventory_depth_standard)

dcant_budget_lat_grid_all <- dcant_budget_lat_grid_all %>% 
  filter(inv_depth == params_global$inventory_depth_standard)

dcant_budget_lon_grid_all <- dcant_budget_lon_grid_all %>% 
  filter(inv_depth == params_global$inventory_depth_standard)

all_predictors <- c("saltempaouoxygenphosphatenitratesilicate")

params_local_all <- params_local_all %>%
  mutate(MLR_predictors = str_remove_all(all_predictors,
                                         MLR_predictors))

dcant_budget_lat_grid_all <- dcant_budget_lat_grid_all %>%
  pivot_wider(names_from = estimate,
              values_from = value) %>%
  filter(period != "1994 - 2014",
         method == "total")

dcant_budget_lon_grid_all <- dcant_budget_lon_grid_all %>%
  pivot_wider(names_from = estimate,
              values_from = value) %>%
  filter(period != "1994 - 2014",
         method == "total")

2 Uncertainty limit

sd_uncertainty_limit <- 4

3 Individual cases

3.1 Absoulte values

dcant_inv_all %>%
  filter(#data_source %in% c("mod", "obs"),
         period != "1994 - 2014") %>%
  group_by(data_source) %>%
  group_split() %>%
  # head(1) %>%
  map(
    ~ p_map_cant_inv(df = .x,
                     var = "dcant",
                     subtitle_text = paste("data_source:",
                                           unique(.x$data_source))) +
      facet_grid(MLR_basins ~ period) +
      theme(axis.text = element_blank(),
            axis.ticks = element_blank())
  )

[[1]]

Past versions of cases_absolute-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

[[2]]

Past versions of cases_absolute-2.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

[[3]]

Past versions of cases_absolute-3.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

dcant_inv_all %>%
  filter(data_source %in% c("mod", "obs"),
         period == "1994 - 2014") %>%
  p_map_cant_inv(
    var = "dcant",
    subtitle_text = "1994 - 2014",
    breaks = c(-Inf, seq(0, 16, 2), Inf) * 2
  ) +
  facet_grid(MLR_basins ~ data_source) +
  theme(axis.text = element_blank(),
        axis.ticks = element_blank())

Past versions of cases_absolute-4.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

p_dcant_inv_all_1994_2004 <-
  dcant_inv_all %>%
  filter(data_source %in% c("obs"),
         period == "1994 - 2004") %>%
  mutate(period = recode(period,
                         "1994 - 2004" = "Observation-based")) %>%
  p_map_cant_inv(var = "dcant",
                 title_text = "1994 - 2004") +
  facet_grid(MLR_basins ~ period) +
  theme(
    axis.text = element_blank(),
    axis.ticks = element_blank(),
    strip.background.y = element_blank(),
    strip.text.y = element_blank(),
    legend.position = "left"
  )

3.2 Biases

dcant_inv_bias_all %>%
  filter(period != "1994 - 2014") %>%
  p_map_cant_inv(var = "dcant_bias",
                 col = "bias",
                 subtitle_text = "data_source: mod - mod_truth") +
  facet_grid(MLR_basins ~ period)

Past versions of cases_bias-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

dcant_inv_bias_all %>%
  filter(period == "1994 - 2014") %>%
  p_map_cant_inv(var = "dcant_bias",
                 col = "bias",
                 subtitle_text = "data_source: mod - mod_truth") +
  facet_grid(MLR_basins ~ period)

Past versions of cases_bias-2.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

p_dcant_inv_bias_all_1994_2004 <-
  dcant_inv_bias_all %>%
  filter(period == "1994 - 2004") %>%
  mutate(period = recode(period,
                         "1994 - 2004" = "Model-based")) %>%
  p_map_cant_inv(var = "dcant_bias",
                 col = "bias",
                 title_text = "Models") +
  facet_grid(MLR_basins ~ period) +
  theme(
    axis.text = element_blank(),
    axis.ticks = element_blank(),
    plot.title = element_blank()
  )

p_dcant_1994_2004 <- 
p_dcant_inv_all_1994_2004 +
  p_dcant_inv_bias_all_1994_2004

p_dcant_1994_2004

Past versions of composed_absolute_and_bias_figure-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

# ggsave(plot = p_dcant_1994_2004,
#        path = "output/other",
#        filename = "inv_dcant_1994_2004_abs_bias.png",
#        height = 8,
#        width = 8)

3.3 Decadal offset

dcant_inv_all %>%
  filter(period != "1994 - 2014") %>%
  select(data_source, lon, lat, MLR_basins, period, dcant_pos) %>% 
  pivot_wider(values_from = dcant_pos,
              names_from = period) %>% 
  mutate(dcant_pos_bias = `2004 - 2014`- `1994 - 2004`) %>% 
  group_by(data_source) %>%
  group_split() %>%
  # head(1) %>%
  map(
    ~ p_map_cant_inv(df = .x,
                     var = "dcant_pos_bias",
                     col = "bias",
                     subtitle_text = paste("data_source:",
                                           unique(.x$data_source))) +
      facet_grid(MLR_basins ~ .) +
      theme(axis.text = element_blank(),
            axis.ticks = element_blank())
  )

[[1]]

Past versions of decadal_dcant_offset_column_inventory_map-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

[[2]]

Past versions of decadal_dcant_offset_column_inventory_map-2.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

[[3]]

Past versions of decadal_dcant_offset_column_inventory_map-3.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

dcant_inv_bias_all %>%
  filter(period != "1994 - 2014") %>%
  select(lon, lat, MLR_basins, period, dcant_pos_bias) %>%
  pivot_wider(values_from = dcant_pos_bias,
              names_from = period) %>%
  mutate(dcant_pos_bias = `2004 - 2014` - `1994 - 2004`) %>%
  p_map_cant_inv(var = "dcant_pos_bias",
                 col = "bias") +
  facet_grid(MLR_basins ~ .) +
  theme(axis.text = element_blank(),
        axis.ticks = element_blank())

Past versions of decadal_bias_offset_column_inventory_map-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

3.3.1 Density distribution

dcant_inv_bias_all %>%
  filter(abs(dcant_bias) < 10) %>% 
  ggplot(aes(dcant_bias, col = MLR_basins)) +
  scale_color_brewer(palette = "Dark2") +
  geom_vline(xintercept = 0) +
  geom_density() +
  facet_grid(period ~.)

Past versions of bias_density_distribution-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

3.4 Lat grid budgets

dcant_budget_lat_grid_all %>%
  group_split(data_source) %>%
  # head(1) %>%
  map(
    ~ ggplot(data = .x,
             aes(lat_grid, dcant, fill = MLR_basins)) +
      geom_hline(yintercept = 0) +
      geom_col(position = "dodge") +
      coord_flip() +
      scale_fill_brewer(palette = "Dark2") +
      labs(title = paste("data_source:", unique(.x$data_source))) +
      facet_grid(basin_AIP ~ period)
  )

[[1]]

Past versions of lat_grid_budget_all-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

[[2]]

Past versions of lat_grid_budget_all-2.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

3.5 Lon grid budgets

dcant_budget_lon_grid_all %>%
  group_split(data_source, period) %>%
  # head(1) %>%
  map(
    ~ ggplot(data = .x,
             aes(lon_grid, dcant, fill = MLR_basins)) +
      geom_col(position = "dodge") +
      scale_fill_brewer(palette = "Dark2") +
      labs(title = paste(
        "data_source:",
        unique(.x$data_source),
        "| period:",
        unique(.x$period)
      )) +
      facet_grid(basin_AIP ~ .)
  )

[[1]]

Past versions of lon_grid_budget_all-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

[[2]]

Past versions of lon_grid_budget_all-2.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

[[3]]

Past versions of lon_grid_budget_all-3.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

[[4]]

Past versions of lon_grid_budget_all-4.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

4 Ensemble

dcant_inv_ensemble <- dcant_inv_all %>% 
  filter(data_source %in% c("mod", "obs")) %>% 
  group_by(lat, lon, data_source, period) %>% 
  summarise(dcant_mean = mean(dcant),
            dcant_sd = sd(dcant),
            dcant_range = max(dcant)- min(dcant)) %>% 
  ungroup()

`summarise()` has grouped output by 'lat', 'lon', 'data_source'. You can
override using the `.groups` argument.

dcant_lat_grid_ensemble <- dcant_budget_lat_grid_all %>% 
  filter(data_source %in% c("mod", "obs")) %>% 
  group_by(lat_grid, data_source, period, MLR_basins) %>% 
  summarise(dcant = sum(dcant, na.rm = TRUE)) %>% 
  ungroup() %>% 
  group_by(lat_grid, data_source, period) %>% 
  summarise(dcant_mean = mean(dcant),
            dcant_sd = sd(dcant),
            dcant_max = max(dcant),
            dcant_min = min(dcant)) %>% 
  ungroup()

`summarise()` has grouped output by 'lat_grid', 'data_source', 'period'. You can override using the `.groups` argument.
`summarise()` has grouped output by 'lat_grid', 'data_source'. You can override using the `.groups` argument.

dcant_lon_grid_ensemble <- dcant_budget_lon_grid_all %>% 
  filter(data_source %in% c("mod", "obs")) %>% 
  group_by(lon_grid, data_source, period, MLR_basins) %>% 
  summarise(dcant = sum(dcant, na.rm = TRUE)) %>% 
  ungroup() %>% 
  group_by(lon_grid, data_source, period) %>% 
  summarise(dcant_mean = mean(dcant),
            dcant_sd = sd(dcant),
            dcant_max = max(dcant),
            dcant_min = min(dcant)) %>% 
  ungroup()

`summarise()` has grouped output by 'lon_grid', 'data_source', 'period'. You can override using the `.groups` argument.
`summarise()` has grouped output by 'lon_grid', 'data_source'. You can override using the `.groups` argument.

4.1 Mean

p_map_cant_inv(
  df = dcant_inv_ensemble %>%
    filter(period != "1994 - 2014",
           data_source == "obs"),
  var = "dcant_mean",
  subtitle_text = paste("Ensemble mean")
) +
  facet_grid(period ~ .) +
  theme(axis.text = element_blank(),
        axis.ticks = element_blank())

Past versions of ensemble_mean-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

p_map_cant_inv(
  df = dcant_inv_ensemble %>%
    filter(data_source == "obs",
           period == "1994 - 2014"),
  var = "dcant_mean",
  title_text = paste("Ensemble mean"),
  breaks = c(-Inf, seq(0, 28, 4), Inf)
) +
  facet_grid(period ~ .) +
  theme(
    axis.text = element_blank(),
    axis.ticks = element_blank(),
    legend.position = "left"
  )

Past versions of ensemble_mean_two_decades-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

4.2 Mean bias

dcant_inv_ensemble_bias <- full_join(
  dcant_inv_ensemble %>%
    filter(data_source == "mod") %>%
    select(lat, lon, period, dcant_mean, dcant_sd),
  dcant_inv_all %>%
    filter(data_source == "mod_truth",
           MLR_basins == unique(dcant_inv_all$MLR_basins)[1]) %>%
    select(lat, lon, period, dcant)
)

Joining, by = c("lat", "lon", "period")

dcant_inv_ensemble_bias <- dcant_inv_ensemble_bias %>%
  mutate(dcant_mean_bias = dcant_mean - dcant)

dcant_inv_ensemble_bias %>%
  filter(period != "1994 - 2014") %>%
  p_map_cant_inv(var = "dcant_mean_bias",
                 col = "bias",
                 subtitle_text = "Ensemble mean - mod_truth") +
  facet_grid(period ~ .)

Past versions of ensemble_mean_bias-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

4.2.1 Density distribution

dcant_inv_bias_all %>%
  ggplot() +
  scale_color_manual(values = c("red", "grey")) +
  geom_vline(xintercept = 0) +
  geom_density(aes(dcant_bias, group = MLR_basins, col = "Individual")) +
  geom_density(data = dcant_inv_ensemble_bias,
               aes(dcant_mean_bias, col = "Ensemble")) +
  facet_grid(period ~.) +
  coord_cartesian(xlim = c(-10, 10))

Past versions of bias_density_distribution_ensemble-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

4.3 Mean lat grid budgets

dcant_lat_grid_ensemble %>%
  ggplot(aes(lat_grid, dcant_mean)) +
  geom_hline(yintercept = 0) +
  geom_col(position = "dodge",
           fill = "grey80",
           col = "grey20") +
  geom_errorbar(aes(
    ymin = dcant_min,
    ymax = dcant_max
  ),
  col = "grey20",
  width = 0) +
  scale_color_brewer(palette = "Set1") +
  coord_flip() +
  scale_fill_brewer(palette = "Dark2") +
  facet_grid(data_source ~ period)

Past versions of lat_grid_budget_ensemble-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

4.4 Mean lon grid budgets

dcant_lon_grid_ensemble %>%
  ggplot(aes(lon_grid, dcant_mean)) +
  geom_col(position = "dodge",
           fill = "grey80",
           col = "grey20") +
  geom_errorbar(aes(
    ymin = dcant_min,
    ymax = dcant_max
  ),
  col = "grey20",
  width = 0) +
  facet_grid(data_source ~ period)

Past versions of lon_grid_budget_ensemble-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

4.5 Standard deviation

p_map_cant_inv(
  df = dcant_inv_ensemble,
  var = "dcant_sd",
  breaks = c(seq(0, 4, 0.4), Inf),
  subtitle_text = paste("Ensemble SD")
) +
  facet_grid(period ~ data_source)

Past versions of ensemble_sd-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

4.6 SD as uncertainty

uncertainty_grid <- dcant_inv_ensemble %>%
  filter(dcant_sd > sd_uncertainty_limit) %>%
  distinct(lon, lat, data_source, period)

uncertainty_grid <- m_grid_horizontal_coarse(uncertainty_grid) %>%
  distinct(lon_grid, lat_grid, data_source, period)

map +
  geom_point(data =
               uncertainty_grid,
             aes(lon_grid, lat_grid),
             shape = 3) +
  facet_grid(period ~ data_source)

Past versions of ensemble_sd_uncertainty-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

4.7 SD vs abs bias

4.7.1 2D bin

dcant_inv_ensemble_bias %>% 
  ggplot(aes(abs(dcant_mean_bias), dcant_sd)) +
  geom_bin2d() +
  scale_fill_viridis_c() +
  facet_grid(. ~ period)

Past versions of ensemble_sd_vs_bias-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

4.7.2 Density distributions

legend_title = expression(Delta * C[ant]~(mol~m^2))

dcant_density_distribution_bias <- dcant_inv_ensemble_bias %>% 
  select(dcant_mean_bias, period) %>% 
  pivot_longer(dcant_mean_bias,
               names_to = "estimate",
               values_to = "value") %>% 
  mutate(data_source = "mod")

dcant_density_distribution <- dcant_inv_ensemble %>% 
  select(dcant_mean, dcant_sd, period, data_source) %>% 
  pivot_longer(dcant_mean:dcant_sd,
               names_to = "estimate",
               values_to = "value")

dcant_density_distribution <- bind_rows(dcant_density_distribution,
                                        dcant_density_distribution_bias) %>%
  mutate(period = factor(period, c("1994 - 2004", "2004 - 2014", "1994 - 2014")))

p_ensemble_distribution <-
  dcant_density_distribution %>%
  filter(period != "1994 - 2014") %>% 
  mutate(
    data_source = recode(data_source,
                         mod = "Model",
                         obs = "Observations"),
    estimate = recode(
      estimate,
      dcant_mean = "Mean",
      dcant_mean_bias = "Bias",
      dcant_sd = "SD"
    )
  ) %>%
  ggplot(aes(value, fill = estimate, col = estimate)) +
  scale_color_brewer(palette = "Dark2") +
  scale_fill_brewer(palette = "Dark2") +
  geom_density(alpha = 0.2) +
  facet_grid(period ~ data_source) +
  labs(title = "Density distributions of column inventory ensemble properties",
       x = legend_title,
       y = "Density of 1x1° grid cells") +
  coord_cartesian(ylim = c(0, 0.5),
                  xlim = c(-5, 32))

# ggsave(plot = p_ensemble_distribution,
#        path = "output/other",
#        filename = "inv_ensemble_distribution.png",
#        height = 4,
#        width = 8)


p_ensemble_distribution <-
  dcant_density_distribution %>%
  filter(data_source == "obs") %>%
  mutate(
    data_source = recode(data_source,
                         mod = "Model",
                         obs = "Observations"),
    estimate = recode(
      estimate,
      dcant_mean = "Mean",
      dcant_mean_bias = "Bias",
      dcant_sd = "SD"
    )
  ) %>%
  ggplot(aes(value, fill = estimate, col = estimate)) +
  scale_color_brewer(palette = "Dark2") +
  scale_fill_brewer(palette = "Dark2") +
  geom_density(alpha = 0.2) +
  facet_grid(period ~ data_source) +
  labs(title = "Density distributions of column inventory ensemble properties",
       x = legend_title,
       y = "Density of 1x1° grid cells") +
  coord_cartesian(ylim = c(0, 0.5),
                  xlim = c(-5, 32))

p_ensemble_distribution

Past versions of density_distributions-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

# ggsave(plot = p_ensemble_distribution,
#        path = "output/other",
#        filename = "inv_ensemble_distribution_all.png",
#        height = 6,
#        width = 5)

4.8 Range

p_map_cant_inv(
  df = dcant_inv_ensemble,
  var = "dcant_range",
  breaks = c(seq(0,8,0.8), Inf),
  subtitle_text = paste("Ensemble range")
) +
  facet_grid(period ~ data_source)

Past versions of ensemble_range-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

4.9 Composed figure

uncertainty_grid <- uncertainty_grid %>% 
  filter(data_source == "obs")

p_map_94 <- p_map_cant_inv(
  df = dcant_inv_ensemble %>%
    filter(period == "1994 - 2004",
           data_source == "obs"),
  var = "dcant_mean",
  subtitle_text = NULL,
  title_text = NULL
) +
  geom_point(
    data =
      uncertainty_grid %>% filter(period == "1994 - 2004"),
    aes(lon_grid, lat_grid),
    shape = 3,
    col = "red"
  ) +
  theme(
    axis.text = element_blank(),
    axis.ticks = element_blank(),
    legend.position = "left"
  ) +
  facet_grid(period ~ .,
             switch = "y")

p_map_04 <- p_map_cant_inv(
  df = dcant_inv_ensemble %>%
    filter(period == "2004 - 2014",
           data_source == "obs"),
  var = "dcant_mean",
  subtitle_text = NULL,
  title_text = NULL
) +
  geom_point(
    data =
      uncertainty_grid %>% filter(period == "2004 - 2014"),
    aes(lon_grid, lat_grid),
    shape = 3,
    col = "red"
  ) +
  theme(
    axis.text = element_blank(),
    axis.ticks = element_blank(),
    legend.position = "left"
  ) +
  facet_grid(period ~ .,
             switch = "y")

p_lon_94 <- dcant_lon_grid_ensemble %>%
  filter(data_source == "obs",
         period == "1994 - 2004") %>%
  ggplot(aes(lon_grid, dcant_mean)) +
  geom_col(position = "dodge",
           fill = "grey80",
           col = "grey20") +
  geom_errorbar(aes(ymin = dcant_min,
                    ymax = dcant_max),
                col = "grey20",
                width = 0) +
  scale_y_continuous(
    limits = c(0, 1),
    expand = c(0, 0),
    name = expression(Delta * C[ant] ~ (PgC)),
    position = "right"
  ) +
  scale_x_continuous(name = "Longitude (°E)",
                     limits = c(20, 380),
                     expand = c(0, 0)) +
  theme(
    axis.title.x = element_blank(),
    axis.text.x = element_blank(),
    axis.ticks.x = element_blank()
  )
# coord_fixed(ratio = 100)

p_lon_04 <- dcant_lon_grid_ensemble %>%
  filter(data_source == "obs",
         period == "2004 - 2014") %>% 
  ggplot(aes(lon_grid, dcant_mean)) +
  geom_col(position = "dodge",
           fill = "grey80",
           col = "grey20") +
  geom_errorbar(aes(
    ymin = dcant_min,
    ymax = dcant_max
  ),
  col = "grey20",
  width = 0) +
  scale_y_continuous(limits = c(0,1), expand = c(0,0),
                     name = expression(Delta*C[ant]~(PgC)),
                     position = "right") +
  scale_x_continuous(name = "Longitude (°E)",
                     limits = c(20,380), expand = c(0,0)) +
  theme(title = element_blank())
  # coord_fixed(ratio = 100)

p_lat_94 <- dcant_lat_grid_ensemble %>%
  filter(data_source == "obs",
         period == "1994 - 2004") %>%
  ggplot(aes(dcant_mean, lat_grid)) +
  geom_col(
    position = "dodge",
    fill = "grey80",
    col = "grey20",
    orientation = "y"
  ) +
  geom_errorbar(aes(xmin = dcant_min,
                    xmax = dcant_max),
                col = "grey20",
                width = 0) +
  scale_x_continuous(
    limits = c(0, 2.8),
    expand = c(0, 0),
    name = expression(Delta * C[ant] ~ (PgC))
  ) +
  scale_y_continuous(
    name = "Latitude (°N)",
    limits = c(-80, 65),
    expand = c(0, 0),
    position = "right"
  ) +
  theme(
    axis.title.x = element_blank(),
    axis.text.x = element_blank(),
    axis.ticks.x = element_blank()
  ) 
  # coord_fixed(ratio = 5e-2)

p_lat_04 <- dcant_lat_grid_ensemble %>%
  filter(data_source == "obs",
         period == "2004 - 2014") %>%
  ggplot(aes(dcant_mean, lat_grid)) +
  geom_col(position = "dodge",
           fill = "grey80",
           col = "grey20",
           orientation = "y") +
  geom_errorbar(aes(xmin = dcant_min,
                    xmax = dcant_max),
                col = "grey20",
                width = 0) +
  scale_x_continuous(
    limits = c(0, 2.8),
    expand = c(0, 0),
    name = expression(Delta * C[ant] ~ (PgC))
  ) +
  scale_y_continuous(name = "Latitude (°N)",
                     limits = c(-80, 65),
                     expand = c(0, 0),
                     position = "right")

t_94 <- grid::textGrob("1994 -\n2004")
t_04 <- grid::textGrob("2004 -\n2014")


layout <- "
BBB#
AAAD
AAAD
CCCE
CCCE
GGG#
"

wrap_plots(B = p_lon_94, A = p_map_94, D = p_lat_94,
           G = p_lon_04, C = p_map_04, E = p_lat_04,
           design = layout,
           guides = "collect")&
  theme(legend.position = "left")

Past versions of composed_figure-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

# ggsave("output/publication/Fig_column_inventories.png",
#        width=11,
#        height=9.27)

5 Cases vs ensemble

5.1 Offset from mean

dcant_inv_all <- full_join(dcant_inv_all,
                           dcant_inv_ensemble)

Joining, by = c("data_source", "lon", "lat", "period")

dcant_inv_all <- dcant_inv_all %>% 
  mutate(dcant_offset = dcant - dcant_mean)

dcant_inv_all %>%
  filter(data_source %in% c("mod", "obs")) %>%
  group_by(period) %>%
  group_split() %>%
  # head(1) %>%
  map(
    ~ p_map_cant_inv(df = .x,
                     var = "dcant_offset",
                     col = "bias",
                     subtitle_text = paste("period:",
                                           unique(.x$period))) +
      facet_grid(MLR_basins ~ data_source)
  )

[[1]]

Past versions of ensemble_deviation_from_mean-1.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

[[2]]

Past versions of ensemble_deviation_from_mean-2.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

[[3]]

Past versions of ensemble_deviation_from_mean-3.png

Version	Author	Date
cc337dd	jens-daniel-mueller	2022-11-11

Session information

sessionInfo()

R version 4.1.2 (2021-11-01)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: openSUSE Leap 15.3

Matrix products: default
BLAS:   /usr/local/R-4.1.2/lib64/R/lib/libRblas.so
LAPACK: /usr/local/R-4.1.2/lib64/R/lib/libRlapack.so

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] geomtextpath_0.1.0 colorspace_2.0-2   marelac_2.1.10     shape_1.4.6       
 [5] ggforce_0.3.3      metR_0.11.0        scico_1.3.0        patchwork_1.1.1   
 [9] collapse_1.7.0     forcats_0.5.1      stringr_1.4.0      dplyr_1.0.7       
[13] purrr_0.3.4        readr_2.1.1        tidyr_1.1.4        tibble_3.1.6      
[17] ggplot2_3.3.5      tidyverse_1.3.1    workflowr_1.7.0   

loaded via a namespace (and not attached):
 [1] fs_1.5.2           bit64_4.0.5        lubridate_1.8.0    gsw_1.0-6         
 [5] RColorBrewer_1.1-2 httr_1.4.2         rprojroot_2.0.2    tools_4.1.2       
 [9] backports_1.4.1    bslib_0.3.1        utf8_1.2.2         R6_2.5.1          
[13] DBI_1.1.2          withr_2.4.3        tidyselect_1.1.1   processx_3.5.2    
[17] bit_4.0.4          compiler_4.1.2     git2r_0.29.0       textshaping_0.3.6 
[21] cli_3.1.1          rvest_1.0.2        xml2_1.3.3         labeling_0.4.2    
[25] sass_0.4.0         scales_1.1.1       checkmate_2.0.0    SolveSAPHE_2.1.0  
[29] callr_3.7.0        systemfonts_1.0.3  digest_0.6.29      rmarkdown_2.11    
[33] oce_1.5-0          pkgconfig_2.0.3    htmltools_0.5.2    highr_0.9         
[37] dbplyr_2.1.1       fastmap_1.1.0      rlang_1.0.2        readxl_1.3.1      
[41] rstudioapi_0.13    jquerylib_0.1.4    generics_0.1.1     farver_2.1.0      
[45] jsonlite_1.7.3     vroom_1.5.7        magrittr_2.0.1     Rcpp_1.0.8        
[49] munsell_0.5.0      fansi_1.0.2        lifecycle_1.0.1    stringi_1.7.6     
[53] whisker_0.4        yaml_2.2.1         MASS_7.3-55        grid_4.1.2        
[57] parallel_4.1.2     promises_1.2.0.1   crayon_1.4.2       haven_2.4.3       
[61] hms_1.1.1          seacarb_3.3.0      knitr_1.37         ps_1.6.0          
[65] pillar_1.6.4       reprex_2.0.1       glue_1.6.0         evaluate_0.14     
[69] getPass_0.2-2      data.table_1.14.2  modelr_0.1.8       vctrs_0.3.8       
[73] tzdb_0.2.0         tweenr_1.0.2       httpuv_1.6.5       cellranger_1.1.0  
[77] gtable_0.3.0       polyclip_1.10-0    assertthat_0.2.1   xfun_0.29         
[81] broom_0.7.11       later_1.3.0        viridisLite_0.4.0  ellipsis_0.3.2    
[85] here_1.0.1