Last updated: 2021-08-09

Checks: 7 0

Knit directory: emlr_obs_v_XXX/

This reproducible R Markdown analysis was created with workflowr (version 1.6.2). 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(20200707)

The command set.seed(20200707) 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: dd6419c

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 dd6419c. 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/

Unstaged changes:
    Modified:   code/Workflowr_project_managment.R
    Modified:   data/auxillary/params_local.rds

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/eMLR_data_preparation.Rmd) and HTML (docs/eMLR_data_preparation.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
Rmd	dd6419c	jens-daniel-mueller	2021-08-09	revised global MLR cleaning
html	cd8e0d5	jens-daniel-mueller	2021-08-06	Build site.
html	15773a0	jens-daniel-mueller	2021-08-06	included calculation of revelle factor
Rmd	d694e17	jens-daniel-mueller	2021-08-06	included calculation of revelle factor
html	da61d1a	jens-daniel-mueller	2021-08-06	Build site.
html	1010198	jens-daniel-mueller	2021-08-06	Build site.
Rmd	3a35523	jens-daniel-mueller	2021-08-06	compare surface obs and equi dcant estimates
html	6da46ae	jens-daniel-mueller	2021-08-06	Build site.
Rmd	5d17b4f	jens-daniel-mueller	2021-08-06	plot C* N vs P for data source
html	340d731	jens-daniel-mueller	2021-08-06	Build site.
html	71546e4	jens-daniel-mueller	2021-08-06	test with stricter CANYON-B filtering
html	29444a1	jens-daniel-mueller	2021-08-05	Build site.
html	42e80c0	jens-daniel-mueller	2021-08-04	Build site.
html	48f6eed	jens-daniel-mueller	2021-08-04	Build site.
html	1c597ab	jens-daniel-mueller	2021-08-04	Build site.
html	81a46a4	jens-daniel-mueller	2021-08-03	Build site.
html	b88c61b	jens-daniel-mueller	2021-08-03	Build site.
html	0f0d5e5	jens-daniel-mueller	2021-08-03	Build site.
html	a53656d	jens-daniel-mueller	2021-08-03	Build site.
html	88f7356	jens-daniel-mueller	2021-08-02	Build site.
html	d759279	jens-daniel-mueller	2021-08-02	Build site.
html	127b801	jens-daniel-mueller	2021-07-24	Build site.
Rmd	414e9ff	jens-daniel-mueller	2021-07-24	test v_201
html	1f65ef1	jens-daniel-mueller	2021-07-23	Build site.
html	912d90e	jens-daniel-mueller	2021-07-23	Build site.
html	2477316	jens-daniel-mueller	2021-07-23	rebuild: surface dcant mapping seperate
html	c9ccc00	jens-daniel-mueller	2021-07-22	Build site.
html	426b2df	jens-daniel-mueller	2021-07-21	Build site.
html	971ce87	jens-daniel-mueller	2021-07-13	Build site.
html	c18a8b1	jens-daniel-mueller	2021-07-09	Build site.
Rmd	795a3b0	jens-daniel-mueller	2021-07-09	complete revision
html	910d64a	jens-daniel-mueller	2021-07-02	Build site.
html	1cbf907	jens-daniel-mueller	2021-07-02	Build site.
html	6f4c8e3	jens-daniel-mueller	2021-06-30	Build site.
Rmd	3be2761	jens-daniel-mueller	2021-06-30	added histogram
html	749d84e	jens-daniel-mueller	2021-06-30	Build site.
Rmd	7fa4dbd	jens-daniel-mueller	2021-06-30	added histogram
html	955f5dc	jens-daniel-mueller	2021-06-29	Build site.
html	9480f94	jens-daniel-mueller	2021-06-29	Build site.
html	d054c39	jens-daniel-mueller	2021-06-24	Build site.
html	87e2739	jens-daniel-mueller	2021-06-24	Build site.
Rmd	6e94317	jens-daniel-mueller	2021-06-24	test coarse grid rarefication
html	4fcd066	jens-daniel-mueller	2021-06-24	Build site.
Rmd	ee48dee	jens-daniel-mueller	2021-06-24	test coarse grid rarefication
html	4e1006c	jens-daniel-mueller	2021-06-18	Build site.
Rmd	9d4550e	jens-daniel-mueller	2021-06-18	cleaned disequilibrium computation
html	3b6387f	jens-daniel-mueller	2021-06-18	Build site.
Rmd	9a6e1b8	jens-daniel-mueller	2021-06-18	rerun with observation specific alpha
html	77cfc43	jens-daniel-mueller	2021-06-17	Build site.
Rmd	3f1ef54	jens-daniel-mueller	2021-06-17	adapt variable naming to delta_pco2_glob
html	3052a6c	jens-daniel-mueller	2021-06-15	Build site.
html	3cb64bd	jens-daniel-mueller	2021-06-15	Build site.
html	c6b3da6	jens-daniel-mueller	2021-06-14	Build site.
Rmd	aeddb8d	jens-daniel-mueller	2021-06-14	corrected Cant calculation at tref2
html	48c73fc	jens-daniel-mueller	2021-06-14	Build site.
html	439ee80	jens-daniel-mueller	2021-06-11	Build site.
html	33ffcab	jens-daniel-mueller	2021-06-10	Build site.
Rmd	ca07f9b	jens-daniel-mueller	2021-06-10	revised sign in disequilibrium ratio
html	7e1f407	jens-daniel-mueller	2021-06-10	Build site.
html	2cbe18c	jens-daniel-mueller	2021-06-10	added zonal mean section control plots
html	ed82630	jens-daniel-mueller	2021-06-10	Build site.
Rmd	fdb6a28	jens-daniel-mueller	2021-06-10	added zonal mean section control plots
html	c5aaa55	jens-daniel-mueller	2021-06-10	Build site.
Rmd	67203cc	jens-daniel-mueller	2021-06-10	added zonal mean section control plots
html	69c79d0	jens-daniel-mueller	2021-06-08	Build site.
Rmd	2a0c856	jens-daniel-mueller	2021-06-08	revised alpha factor for total Cant scaling to tref
html	1772903	jens-daniel-mueller	2021-06-07	Build site.
Rmd	7c32ee8	jens-daniel-mueller	2021-06-07	revised beta factor for total Cant scaling to tref
html	594ed9a	jens-daniel-mueller	2021-06-04	Build site.
html	db7df0e	jens-daniel-mueller	2021-06-04	rebuild without overlapping eras
Rmd	71f97a3	jens-daniel-mueller	2021-06-04	rebuild without overlapping eras
html	2edc791	jens-daniel-mueller	2021-06-04	Build site.
Rmd	64a0608	jens-daniel-mueller	2021-06-04	calculate cant total at tref2 based on alpha
html	207339d	jens-daniel-mueller	2021-06-03	Build site.
html	315710b	jens-daniel-mueller	2021-06-03	include anomalous changes
html	1b625b5	jens-daniel-mueller	2021-06-03	Build site.
Rmd	2433ed8	jens-daniel-mueller	2021-06-03	include anomalous changes
html	be90356	jens-daniel-mueller	2021-06-02	Build site.
html	d37a85d	jens-daniel-mueller	2021-05-31	Build site.
Rmd	1c2aff8	jens-daniel-mueller	2021-05-31	test run with beta version of GLODAPv2.2021
html	4b7a5ee	jens-daniel-mueller	2021-05-28	Build site.
html	12b455a	jens-daniel-mueller	2021-05-27	Build site.
Rmd	ee13efb	jens-daniel-mueller	2021-05-27	optional source of local params fully implemented
html	8c736a6	jens-daniel-mueller	2021-05-27	Build site.
Rmd	dc8e4e1	jens-daniel-mueller	2021-05-27	optional source of local params implemented
html	25bd183	jens-daniel-mueller	2021-05-26	Build site.
html	b79cb2d	jens-daniel-mueller	2021-05-20	Build site.
html	62bd574	jens-daniel-mueller	2021-05-20	Build site.
html	7c56c39	jens-daniel-mueller	2021-05-19	Build site.
Rmd	4a1fd72	jens-daniel-mueller	2021-05-19	test completely overlapping eras
html	0de759e	jens-daniel-mueller	2021-05-13	Build site.
html	52e7583	jens-daniel-mueller	2021-05-12	Build site.
Rmd	5ac5bef	jens-daniel-mueller	2021-05-12	rerun without sea of japan
html	969e631	jens-daniel-mueller	2021-05-12	Build site.
html	d2a83bc	jens-daniel-mueller	2021-04-16	Build site.
html	c0a47df	jens-daniel-mueller	2021-04-16	Build site.
html	50290e8	jens-daniel-mueller	2021-04-16	overlapping eras
html	a00ec94	jens-daniel-mueller	2021-04-16	Build site.
Rmd	82df560	jens-daniel-mueller	2021-04-16	corrected tref assignment
html	b6fe355	jens-daniel-mueller	2021-04-16	Build site.
html	81b7c6d	jens-daniel-mueller	2021-04-16	Build site.
html	ddec5b7	jens-daniel-mueller	2021-04-15	Build site.
html	29edae5	jens-daniel-mueller	2021-04-14	Build site.
html	099d566	jens-daniel-mueller	2021-04-14	Build site.
html	bb44686	jens-daniel-mueller	2021-04-14	Build site.
html	bf40480	jens-daniel-mueller	2021-04-13	Build site.
html	9f31fe3	jens-daniel-mueller	2021-04-13	Build site.
Rmd	80e1ee3	jens-daniel-mueller	2021-04-13	rerun post-2000 with model data
html	c98be84	jens-daniel-mueller	2021-04-13	Build site.
Rmd	1df2496	jens-daniel-mueller	2021-04-13	included model data
html	338dd3c	jens-daniel-mueller	2021-04-09	Build site.
html	a79ca2c	jens-daniel-mueller	2021-04-09	included model data
html	be095c6	jens-daniel-mueller	2021-04-09	Build site.
Rmd	1c2bdb7	jens-daniel-mueller	2021-04-09	included model data
html	7831fd3	jens-daniel-mueller	2021-04-09	Build site.
html	20b26d4	jens-daniel-mueller	2021-04-08	Build site.
html	c0895f8	jens-daniel-mueller	2021-04-07	Build site.
html	156d5b7	jens-daniel-mueller	2021-04-07	Build site.
html	eb827c9	jens-daniel-mueller	2021-04-07	Build site.
Rmd	49be8ed	jens-daniel-mueller	2021-03-26	included model data
html	857bad3	jens-daniel-mueller	2021-03-24	Build site.
Rmd	02e91d4	jens-daniel-mueller	2021-03-24	included model data
html	03b6009	jens-daniel-mueller	2021-03-23	removed before copying template
html	555750f	jens-daniel-mueller	2021-03-23	Build site.
Rmd	2021931	jens-daniel-mueller	2021-03-23	restriction to two eras and new definition procedure
html	f155edd	jens-daniel-mueller	2021-03-23	Build site.
Rmd	6601b92	jens-daniel-mueller	2021-03-23	rerun w/o post 2012 era
html	380d215	jens-daniel-mueller	2021-03-21	Build site.
html	33b385b	jens-daniel-mueller	2021-03-20	Build site.
html	330dcd0	jens-daniel-mueller	2021-03-20	Build site.
html	83a13de	jens-daniel-mueller	2021-03-20	Build site.
html	cf98c6d	jens-daniel-mueller	2021-03-16	Build site.
html	a1d52ff	jens-daniel-mueller	2021-03-15	Build site.
html	0bade3b	jens-daniel-mueller	2021-03-15	Build site.
html	27c1f4b	jens-daniel-mueller	2021-03-14	Build site.
html	af75ebf	jens-daniel-mueller	2021-03-14	Build site.
Rmd	e3dde84	jens-daniel-mueller	2021-03-14	test without filtering
html	5017709	jens-daniel-mueller	2021-03-11	Build site.
Rmd	7a953b4	jens-daniel-mueller	2021-03-11	test with weak filtering
html	585b07f	jens-daniel-mueller	2021-03-11	Build site.
Rmd	f568b8e	jens-daniel-mueller	2021-03-11	cleaned cstar filtering
html	6482ed7	jens-daniel-mueller	2021-03-11	Build site.
html	85a5ed2	jens-daniel-mueller	2021-03-10	Build site.
Rmd	749b5db	jens-daniel-mueller	2021-03-10	filter based on comparison to CANYON-B
html	081a12a	jens-daniel-mueller	2021-03-06	Build site.
Rmd	dafd78e	jens-daniel-mueller	2021-03-06	Canyon-B comparison for singla and all years
html	dd2b3a1	jens-daniel-mueller	2021-03-06	Build site.
Rmd	bc38fd4	jens-daniel-mueller	2021-03-06	Canyon-B comparison for singla and all years
html	1dc27ab	jens-daniel-mueller	2021-03-06	Build site.
Rmd	8abeada	jens-daniel-mueller	2021-03-06	Canyon-B comparison for singla and all years
html	e59c7c3	jens-daniel-mueller	2021-03-06	Build site.
Rmd	c3e762b	jens-daniel-mueller	2021-03-06	Canyon-B comparison for all years together
html	98fb407	jens-daniel-mueller	2021-03-06	Build site.
Rmd	bd24d69	jens-daniel-mueller	2021-03-06	Canyon-B comparison for all years
html	c71fdae	jens-daniel-mueller	2021-03-05	Build site.
Rmd	8a40d10	jens-daniel-mueller	2021-03-05	Canyon-B comparison for all years
html	00688a1	jens-daniel-mueller	2021-03-05	Build site.
Rmd	a805084	jens-daniel-mueller	2021-03-05	comparison to CANYON-B started
html	6c0bec6	jens-daniel-mueller	2021-03-05	Build site.
Rmd	559dd52	jens-daniel-mueller	2021-03-05	rebuild with cruise rmse < 10 filter
html	3c2ec33	jens-daniel-mueller	2021-03-05	Build site.
Rmd	006c875	jens-daniel-mueller	2021-03-05	rebuild with NA in Cant replaced by 0
html	af70b94	jens-daniel-mueller	2021-03-04	Build site.
Rmd	c9cf1fd	jens-daniel-mueller	2021-03-04	rebuild with NA in Cant replaced by 0
html	86406d5	jens-daniel-mueller	2021-02-24	Build site.
html	3d3b4cc	jens-daniel-mueller	2021-02-23	Build site.
html	7b672f7	jens-daniel-mueller	2021-01-11	Build site.
html	33ba23c	jens-daniel-mueller	2021-01-07	Build site.
html	318609d	jens-daniel-mueller	2020-12-23	adapted more variable predictor selection
html	9d0b2d0	jens-daniel-mueller	2020-12-23	Build site.
Rmd	a4531df	jens-daniel-mueller	2020-12-23	test 106
html	0aa2b50	jens-daniel-mueller	2020-12-23	remove html before duplication
html	39113c3	jens-daniel-mueller	2020-12-23	Build site.
Rmd	bef9220	jens-daniel-mueller	2020-12-23	rebuild before sensitivity test
html	2886da0	jens-daniel-mueller	2020-12-19	Build site.
html	02f0ee9	jens-daniel-mueller	2020-12-18	cleaned up for copying template
html	965dba3	jens-daniel-mueller	2020-12-18	Build site.
html	5d452fe	jens-daniel-mueller	2020-12-18	Build site.
Rmd	ca65bf5	jens-daniel-mueller	2020-12-18	rebuild after final cleaning
html	7bcb4eb	jens-daniel-mueller	2020-12-18	Build site.
Rmd	cd02e63	jens-daniel-mueller	2020-12-18	rebuild with new reference year adjustment
html	d397028	jens-daniel-mueller	2020-12-18	Build site.
html	7131186	jens-daniel-mueller	2020-12-17	Build site.
html	22b07fb	jens-daniel-mueller	2020-12-17	Build site.
html	f3a708f	jens-daniel-mueller	2020-12-17	Build site.
html	e4ca289	jens-daniel-mueller	2020-12-16	Build site.
html	158fe26	jens-daniel-mueller	2020-12-15	Build site.
Rmd	449195a	jens-daniel-mueller	2020-12-15	rebuild without subsetting nitrate
html	7a9a4cb	jens-daniel-mueller	2020-12-15	Build site.
Rmd	d234226	jens-daniel-mueller	2020-12-15	rebuild with cstar_tref
html	61b263c	jens-daniel-mueller	2020-12-15	Build site.
html	4d612dd	jens-daniel-mueller	2020-12-15	Build site.
Rmd	e7e5ff1	jens-daniel-mueller	2020-12-15	rebuild with eMLR target variable selection
html	953caf3	jens-daniel-mueller	2020-12-15	Build site.
html	42daf5c	jens-daniel-mueller	2020-12-14	Build site.
Rmd	923aa7f	jens-daniel-mueller	2020-12-14	rebuild with new path and auto folder creation
html	984697e	jens-daniel-mueller	2020-12-12	Build site.
html	3ebff89	jens-daniel-mueller	2020-12-12	Build site.
html	7d82772	jens-daniel-mueller	2020-12-11	Build site.
Rmd	6069c23	jens-daniel-mueller	2020-12-11	selectable basinmask, try 5
html	7788175	jens-daniel-mueller	2020-12-09	Build site.
Rmd	64b795c	jens-daniel-mueller	2020-12-09	added histograms after data preparation
html	b01a367	jens-daniel-mueller	2020-12-09	Build site.
html	24a632f	jens-daniel-mueller	2020-12-07	Build site.
html	92dca91	jens-daniel-mueller	2020-12-07	Build site.
html	6a8004b	jens-daniel-mueller	2020-12-07	Build site.
html	70bf1a5	jens-daniel-mueller	2020-12-07	Build site.
html	7555355	jens-daniel-mueller	2020-12-07	Build site.
html	143d6fa	jens-daniel-mueller	2020-12-07	Build site.
html	3c8a83c	jens-daniel-mueller	2020-12-04	Build site.
Rmd	e11f455	jens-daniel-mueller	2020-12-04	improved output plots by using stat_2d instead of points
html	abc6818	jens-daniel-mueller	2020-12-03	Build site.
Rmd	992ba15	jens-daniel-mueller	2020-12-03	rebuild with variable inventory depth
html	090e4d5	jens-daniel-mueller	2020-12-02	Build site.
Rmd	c98d27b	jens-daniel-mueller	2020-12-02	cleaned subsetting and data preparation
html	37e9dac	jens-daniel-mueller	2020-12-02	Build site.
Rmd	9ff071b	jens-daniel-mueller	2020-12-02	minor improvement of tref adejustment, formatting
html	7c25f7a	jens-daniel-mueller	2020-12-02	Build site.
html	ec8dc38	jens-daniel-mueller	2020-12-02	Build site.
html	c987de1	jens-daniel-mueller	2020-12-02	Build site.
html	f8358f8	jens-daniel-mueller	2020-12-02	Build site.
html	b03ddb8	jens-daniel-mueller	2020-12-02	Build site.
Rmd	9183e8f	jens-daniel-mueller	2020-12-02	revised assignment of era to eras
html	22d0127	jens-daniel-mueller	2020-12-01	Build site.
html	0ff728b	jens-daniel-mueller	2020-12-01	Build site.
html	b02b7a4	jens-daniel-mueller	2020-12-01	Build site.
Rmd	60bea48	jens-daniel-mueller	2020-12-01	auto eras naming
html	cf19652	jens-daniel-mueller	2020-11-30	Build site.
Rmd	2842970	jens-daniel-mueller	2020-11-30	cleaned for eMLR part only
html	196be51	jens-daniel-mueller	2020-11-30	Build site.
Rmd	7a4b015	jens-daniel-mueller	2020-11-30	first rebuild on ETH server
Rmd	bc61ce3	Jens Müller	2020-11-30	Initial commit
html	bc61ce3	Jens Müller	2020-11-30	Initial commit

1 Version ID

The results displayed on this site correspond to the Version_ID:

params$Version_ID

[1] "v_XXX"

2 Required data

2.1 GLODAP

GLODAP <-
  read_csv(paste(path_version_data,
                 "GLODAPv2.2020_clean.csv",
                 sep = ""))

2.2 Cant fields

S04_cant_3d <-
  read_csv(paste(path_preprocessing,
                 "S04_tcant_3d.csv",
                 sep = ""))

G19_cant_3d <-
  read_csv(paste(path_preprocessing,
                 "G19_dcant_3d.csv",
                 sep = ""))

m94_cant_3d <-
  read_csv(paste(
    path_preprocessing_model,
    "cant_annual_field_AD/cant_1994.csv",
    sep = ""
  ))

m07_cant_3d <-
  read_csv(paste(
    path_preprocessing_model,
    "cant_annual_field_AD/cant_2007.csv",
    sep = ""
  ))

2.3 CO2 atm and disequi

co2_atm <-
  read_csv(paste(path_preprocessing,
                 "co2_atm.csv",
                 sep = ""))

co2_atm_reccap2 <-
  read_csv(paste(path_preprocessing,
                 "co2_atm_reccap2.csv",
                 sep = ""))

delta_pco2_annual_mod <-
    read_csv(
      file = paste(
        path_preprocessing_model,
        "surface_ocean_disequilibrium/",
        "C",
        "_annual.csv",
        sep = ""
      )
    )

delta_pco2_annual_obs <-
    read_csv(
      file = paste(
        path_preprocessing,
        "OceanSODA_disequilibrium_annual.csv",
        sep = ""
      )
    )

delta_pco2_annual <- bind_rows(
  delta_pco2_annual_mod %>% mutate(data_source = "mod"),
  delta_pco2_annual_obs %>% mutate(data_source = "obs")
)

rm(delta_pco2_annual_mod, delta_pco2_annual_obs)

2.4 Surface ocean

### model data

variables <-
  c("talkos", "spco2", "tos", "sos")

for (i_variable in variables) {
  
  temp <- read_csv(paste0(
    path_root,
    "/model/preprocessing/surface_ocean_A/", i_variable, ".csv")
  )
  
  if (exists("surface_mod")) {
    surface_mod <- full_join(surface_mod, temp)
  }
  
  if (!exists("surface_mod")) {
    surface_mod <- temp
  }
}

rm(temp, i_variable, variables)

surface_mod <- surface_mod %>% 
  rename(talk = talkos,
         pCO2 = spco2,
         temp = tos,
         sal = sos)

### Observational data (OceanSODA)

surface_obs <- read_csv(paste0(
    path_observations,
    "preprocessing/OceanSODA.csv")
  )

surface_obs <- surface_obs %>% 
  select(lon, lat, year, talk, pCO2, sal, temp)

surface <- bind_rows(
  surface_obs %>% mutate(data_source = "obs"),
  surface_mod %>% mutate(data_source = "mod")
)

rm(surface_mod, surface_obs)

2.5 Others

tref <-
  read_csv(paste(path_version_data,
                 "tref.csv",
                 sep = ""))

2.6 Convert to long format

GLODAP <- GLODAP %>% 
  rename_with(~ gsub("_model", "_mod", .x)) %>% 
  rename_with(.cols = c(temp, sal, gamma, tco2, talk, phosphate,
                        oxygen, aou, nitrate, silicate),
              ~ paste(.x, "obs", sep = "_"))

GLODAP <- GLODAP %>% 
  pivot_longer(
    -c(year:depth), 
    names_to = c(".value", "data_source"), 
    names_sep = "_"
  )

3 PO₄* calculation

The predictor PO₄* was be calculated according to Clement and Gruber (2018), ie based on oxygen. Please note that an erroneous equations for PO4* calculation is given in the supplement of Gruber et al (2019), based on nitrate.

Here we use following equation, if PO₄* is used as predictor:

print(b_phosphate_star)

function (phosphate, oxygen) 
{
    phosphate_star = phosphate + (oxygen/params_local$rPO) - 
        params_local$rPO_offset
    return(phosphate_star)
}

if ("phosphate_star" %in% params_local$MLR_predictors) {
GLODAP <- GLODAP %>% 
  mutate(phosphate_star = b_phosphate_star(phosphate, oxygen))
}

Note, currently we decided to avoid using PO₄* as predictor, because it is equivalent to using PO₄ and O₂.

4 C*

C* serves as a conservative tracer of anthropogenic CO₂ uptake. It is derived from measured DIC by removing the impact of

organic matter formation and respiration
calcification and calcium carbonate dissolution

Contributions of those processes are estimated from phosphate and alkalinity concentrations.

4.1 Stoichiometric ratios

The stoichiometric nutrient ratios for the production and mineralization of organic matter were set to:

C/P: 117
N/P: 16

4.2 Calculation

C* can optionally be calculated based on phosphate or nitrate.

GLODAP <- GLODAP %>%
  mutate(
    cstar_nitrate = b_cstar_nitrate(tco2, nitrate, talk),
    cstar_phosphate = b_cstar_phosphate(tco2, phosphate, talk)
  )


p_prop_prop(df = GLODAP,
            var1 = "cstar_phosphate",
            var2 = "cstar_nitrate") +
  facet_grid(data_source ~ basin_AIP)

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
6da46ae	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
29444a1	jens-daniel-mueller	2021-08-05
42e80c0	jens-daniel-mueller	2021-08-04
48f6eed	jens-daniel-mueller	2021-08-04
81a46a4	jens-daniel-mueller	2021-08-03
b88c61b	jens-daniel-mueller	2021-08-03
88f7356	jens-daniel-mueller	2021-08-02
d759279	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c9ccc00	jens-daniel-mueller	2021-07-22
971ce87	jens-daniel-mueller	2021-07-13
c18a8b1	jens-daniel-mueller	2021-07-09

C* was actually calculated as:

if (params_local$cstar_nutrient == "phosphate") {
  print(b_cstar_phosphate)
  
  GLODAP <- GLODAP %>%
    mutate(cstar_phosphate = b_cstar_phosphate(tco2, phosphate, talk)) %>%
    rename(cstar = cstar_phosphate) %>%
    select(-cstar_nitrate)
}

if (params_local$cstar_nutrient == "nitrate") {
  print(b_cstar_nitrate)
  
  GLODAP <- GLODAP %>%
    mutate(cstar_nitrate = b_cstar_nitrate(tco2, nitrate, talk)) %>%
    rename(cstar = cstar_nitrate) %>%
    select(-cstar_phosphate)
}

function (tco2, nitrate, talk) 
{
    cstar = tco2 - (params_local$rCP/params_local$rNP * nitrate) - 
        0.5 * (talk - nitrate)
    return(cstar)
}

4.3 Reference year adjustment

To adjust observation-based C* values to the reference year of each observation period, we assume a transient steady state change of cant between the time of sampling the reference year. The adjustment requires an approximation of the total cant concentration at the reference year. We approximate this concentration for tref(1) by adding the delta cant signal estimated by Gruber et al (2019) to the “base line” total cant concentration determined for 1994 by Sabine et al (2004):

Cant(tref) = S04 + (tref-1994)/13 * G19

This way, we use exactly S04+G19 for tref(1)=2007. For all other tref(1) we scale Cant with the observed anomalous change over the 1994-2007 period, rather than assuming a transient steady state. However, one assumes a linear behaviour of the anomalous change over time, which might be wrong in particular for the years past 2007.

Therefore, we estimate the total Cant at tref(2) by scaling total cant at tref(1) with alpha, according to Gruber et al. (2019).

For the model data, we perform an analogous adjustment based on the total Cant estimate for 1994 (corresponding to Sabine 2004), and the delta Cant estimate for 1994 - 2007 (corresponding to Gruber 2019).

4.3.1 Join Cant fields

Join observation-based Cant fields of G19 and S04

G19_cant_3d <- G19_cant_3d %>% 
  select(lon, lat, depth, cant_pos_G19 = dcant_pos)

S04_cant_3d <- S04_cant_3d %>% 
  select(lon, lat, depth, cant_pos_S04 = tcant_pos)

# check regional coverage of both data sets
cant_3d_coverage <- full_join(
  S04_cant_3d %>% distinct(lat, lon),
  G19_cant_3d %>% distinct(lat, lon)
)

cant_3d_coverage <- full_join(
  cant_3d_coverage,
  G19_cant_3d %>% distinct(lat, lon) %>% mutate(G19 = "y")
)

cant_3d_coverage <- full_join(
  cant_3d_coverage,
  S04_cant_3d %>% distinct(lat, lon) %>% mutate(S04 = "y")
)

cant_3d_coverage <- cant_3d_coverage %>% 
  mutate(coverage = case_when(
    G19 == "y" & S04 == "y" ~ "both",
    is.na(G19) & S04 == "y" ~ "S04",
    G19 == "y" & is.na(S04) ~ "G19"))

map +
  geom_raster(data = cant_3d_coverage,
              aes(lon, lat, fill = coverage)) +
  geom_raster(data = GLODAP %>% distinct(lat, lon),
              aes(lon, lat)) +
  scale_fill_brewer(palette = "Dark2")

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
29444a1	jens-daniel-mueller	2021-08-05
42e80c0	jens-daniel-mueller	2021-08-04
48f6eed	jens-daniel-mueller	2021-08-04
81a46a4	jens-daniel-mueller	2021-08-03
b88c61b	jens-daniel-mueller	2021-08-03
88f7356	jens-daniel-mueller	2021-08-02
d759279	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c9ccc00	jens-daniel-mueller	2021-07-22
c18a8b1	jens-daniel-mueller	2021-07-09

rm(cant_3d_coverage)


# join data sets
cant_3d <- full_join(S04_cant_3d, G19_cant_3d)

cant_3d <- cant_3d %>%
  mutate(cant_pos_S04 = replace_na(cant_pos_S04, 0),
         cant_pos_G19 = replace_na(cant_pos_G19, 0))

Join model Cant fields of 1994 and 2007

# define positive cant
m94_cant_3d <- m94_cant_3d %>% 
  mutate(cant_pos = if_else(cant_total <= 0, 0, cant_total)) %>% 
  select(lon, lat, depth, cant_pos_S04 = cant_pos)

m07_cant_3d <- m07_cant_3d %>% 
  mutate(cant_pos = if_else(cant_total <= 0, 0, cant_total)) %>% 
  select(lon, lat, depth, cant_pos_m07 = cant_pos)

# join data sets
mod_cant_3d <- full_join(
  m94_cant_3d,
  m07_cant_3d
)

# calculate the delta cant signal, i.e. the analogue to G19
mod_cant_3d <- mod_cant_3d %>% 
  mutate(cant_pos_G19 = cant_pos_m07 - cant_pos_S04,
         cant_pos_G19 = if_else(cant_pos_G19 <= 0, 
                                0, cant_pos_G19)) %>% 
  select(-cant_pos_m07)

rm(m07_cant_3d,
   m94_cant_3d)

Join mod and obs Cant

cant_3d <- bind_rows(
  mod_cant_3d %>% mutate(data_source = "mod"),
  cant_3d %>% mutate(data_source = "obs")
)

cant_3d <- inner_join(cant_3d, basinmask)

4.3.2 tcant at tref1

Calculate total Cant at tref1 by adding G19 to S04, linearly scaled for the time since 1994. Note: For tref1 close to 1994, this should be a marginal change compared to taking SO4 directly.

tcant_3d <- cant_3d %>%
  mutate(tcant_pos = cant_pos_S04 +
           ((min(tref$median_year) - 1994) / 13 * cant_pos_G19))

rm(cant_3d)

# join cant with tref
tcant_3d <- expand_grid(tcant_3d, tref)


# linear scaling of Gruber 2019
# calculate cant fields for all tref
tcant_3d_t1 <- tcant_3d %>%
  filter(median_year == min(tref$median_year))

4.3.3 Revelle factor ratio

# determine mean year for both eras
tref_mean <- c(sort(tref$median_year)[1]-50 , mean(tref$median_year))

# extract surface ocean data for second mean year
surface <- surface %>% 
  filter(year == mean(tref$median_year))

# extract atm pCO2 level for time before first era
delta_pCO2_hist <- co2_atm_reccap2 %>% 
  filter(year %in% tref_mean) %>% 
  arrange(year) %>% 
  mutate(delta_pCO2_hist = pCO2 - lag(pCO2)) %>% 
  drop_na() %>% 
  pull(delta_pCO2_hist)

# derive historic surface pCO2 field
surface_hist <- surface %>% 
  mutate(pCO2 = pCO2 - delta_pCO2_hist,
         year = tref_mean[1])

surface <- bind_rows(
  surface,
  surface_hist
)

map +
  geom_raster(data = surface,
              aes(lon, lat, fill=pCO2)) +
  scale_fill_viridis_c() +
  facet_grid(data_source ~ year)

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06

# calculate surface revelle factor dibution
surface <- surface %>%
  drop_na() %>% 
  mutate(
    rev_fac = buffer(
      flag = 24,
      var1 = pCO2,
      var2 = talk * 1e-6,
      S = sal,
      T = temp,
      P = 0,
      k1k2 = "l"
    )$BetaD
  )

map +
  geom_raster(data = surface,
              aes(lon, lat, fill=rev_fac)) +
  scale_fill_viridis_c() +
  facet_grid(data_source ~ year)

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06

# calculate revelle factor ratios
surface <- surface %>% 
  arrange(year) %>% 
  group_by(lon, lat, data_source) %>% 
  mutate(rev_fac_ratio = lag(rev_fac) / rev_fac) %>% 
  ungroup() %>% 
  drop_na()

map +
  geom_raster(data = surface,
              aes(lon, lat, fill=rev_fac_ratio)) +
  scale_fill_viridis_c() +
  facet_grid(data_source ~ .)

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06

# area weighted mean changes in revelle factor
rev_fac_ratio <- surface %>% 
  mutate(area = earth_surf(lat = lat),
         rev_fac_ratio_scaled = area * rev_fac_ratio) %>% 
  group_by(data_source) %>% 
  summarise(rev_fac_ratio = sum(rev_fac_ratio_scaled) / sum(area)) %>% 
  ungroup()

rm(surface, surface_hist, delta_pCO2_hist, tref_mean)

4.3.4 Alpha

# extract atm pCO2 at reference years
co2_atm_tref <- right_join(co2_atm, tref %>% rename(year = median_year)) %>% 
  select(-year) %>% 
  rename(pCO2_tref = pCO2)

# calculate change in atm pCO2
co2_atm_tref <- co2_atm_tref %>% 
  mutate(delta_pCO2_atm = pCO2_tref - 
           lag(pCO2_tref, default = params_local$preind_atm_pCO2))

# define ratio of changes in revelle factor
# rev_fac_ratio <- 0.92
# print(c("Revelle factor ratio (fixed):", rev_fac_ratio))

# derive change in atm-sea disequilibrium pCO2
delta_pco2_annual <- right_join(delta_pco2_annual,
                                GLODAP %>%
                                  distinct(year, era)) %>%
  drop_na()

# plot disequilibrium time series
delta_pco2_annual %>%
  ggplot(aes(year, delta_pco2_glob)) +
  geom_path(aes(linetype=data_source)) +
  geom_point(aes(col = era, shape=data_source)) +
  scale_color_brewer(palette = "Set1")

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
88f7356	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c18a8b1	jens-daniel-mueller	2021-07-09

# create data frame with preindustrial disequilibrium for model and observations
disequi_pCO2_pi <- bind_cols(
  era = rep(paste0("1750-",tref$median_year[1]),2),
  data_source = c("mod", "obs"),
  mean_delta_pCO2 = c(0,3)
)

disequi_pCO2_tref <- delta_pco2_annual %>% 
  group_by(era, data_source) %>% 
  summarise(mean_delta_pCO2 = mean(delta_pco2_glob)) %>% 
  ungroup()

disequi_pCO2_tref <- bind_rows(
  disequi_pCO2_pi,
  disequi_pCO2_tref
)
rm(disequi_pCO2_pi)

disequi_pCO2_tref <- disequi_pCO2_tref %>% 
  arrange(era) %>% 
  group_by(data_source) %>% 
  mutate(disequi_pCO2_change = mean_delta_pCO2 - lag(mean_delta_pCO2)) %>% 
  ungroup() %>% 
  drop_na()

# calculate xeta ratio
disequi_pCO2_change <- full_join(
  disequi_pCO2_tref,
  co2_atm_tref)

disequi_pCO2_change <- disequi_pCO2_change %>% 
  mutate(xeta = (delta_pCO2_atm - disequi_pCO2_change) / delta_pCO2_atm) %>% 
  arrange(era) %>% 
  group_by(data_source) %>% 
  mutate(xeta_ratio = xeta / lag(xeta),
         delta_pCO2_atm_ratio = delta_pCO2_atm / lag(delta_pCO2_atm)) %>% 
  ungroup()

disequi_pCO2_change <- full_join(
  disequi_pCO2_change,
  rev_fac_ratio
)
rm(rev_fac_ratio)

# determine alpha scaling factor
disequi_pCO2_change <- disequi_pCO2_change %>% 
  mutate(alpha = delta_pCO2_atm_ratio * xeta_ratio * rev_fac_ratio) %>% 
  select(-c(pCO2_tref, delta_pCO2_atm_ratio))

disequi_pCO2_change

# A tibble: 4 x 11
  era   data_source mean_delta_pCO2 disequi_pCO2_ch… start   end delta_pCO2_atm
  <chr> <chr>                 <dbl>            <dbl> <dbl> <dbl>          <dbl>
1 2000… mod                   -8.58            -8.58  2000  2009           96.8
2 2000… obs                   -7.50           -10.5   2000  2009           96.8
3 2010… mod                   -9.86            -1.28  2010  2019           20.3
4 2010… obs                  -10.2             -2.71  2010  2019           20.3
# … with 4 more variables: xeta <dbl>, xeta_ratio <dbl>, rev_fac_ratio <dbl>,
#   alpha <dbl>

alpha_df <- disequi_pCO2_change %>% 
  select(data_source, alpha) %>% 
  drop_na()

4.3.5 tcant at tref2

Calculate total Cant at tref2 by scaling the tref1 estimate with alpha according to Gruber et al 2019.

tcant_3d_t2 <- full_join(tcant_3d %>% filter(median_year == max(tref$median_year)),
                        alpha_df) %>% 
  mutate(tcant_pos = tcant_pos * (1 + alpha)) %>% 
  select(-alpha)

tcant_3d <- bind_rows(tcant_3d_t1, tcant_3d_t2)
rm(tcant_3d_t1, tcant_3d_t2)

# remove columns
tcant_3d <- tcant_3d %>% 
  select(data_source, basin_AIP, lon, lat, depth, era, tcant_pos)

tcant_3d %>% 
  ggplot(aes(tcant_pos, depth)) +
  geom_bin2d() +
  facet_grid(era ~ data_source) +
  scale_fill_viridis_c() +
  scale_y_reverse()

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
88f7356	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c18a8b1	jens-daniel-mueller	2021-07-09

tcant_zonal <- tcant_3d %>%
  group_by(data_source, era) %>%
  nest() %>%
  mutate(section = map(.x = data, ~m_zonal_mean_sd(.x))) %>%
  select(-data) %>%
  unnest(section)

tcant_3d <- tcant_3d %>% 
  select(-basin_AIP)

4.3.6 Combine GLODAP + tcant

# observations grid per era
GLODAP_obs_grid_era <- GLODAP %>% 
  distinct(lat, lon, era, data_source)

# cant data at observations grid
tcant_3d_obs <- left_join(
  GLODAP_obs_grid_era,
  tcant_3d)

tcant_3d_obs <- tcant_3d_obs %>%
  mutate(tcant_pos = replace_na(tcant_pos, 0),
         depth = replace_na(depth, 0))

# calculate number of cant data points per grid cell
tcant_3d_obs <- tcant_3d_obs %>%
  group_by(lon, lat, era, data_source) %>% 
  mutate(n = n(),
         n_group = if_else(n > 1, "n > 1", "n <= 1")) %>% 
  ungroup()

# GLODAP observations with only one Cant value
map +
  geom_raster(data = tcant_3d_obs,
             aes(lon, lat, fill = n_group)) +
  scale_fill_brewer(palette = "Set1", name="n") +
  facet_grid(data_source ~ era) +
  labs(title = "Number of Cant depth levels",
       subtitle = "available per latxlon grid cell")

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
29444a1	jens-daniel-mueller	2021-08-05
42e80c0	jens-daniel-mueller	2021-08-04
48f6eed	jens-daniel-mueller	2021-08-04
81a46a4	jens-daniel-mueller	2021-08-03
b88c61b	jens-daniel-mueller	2021-08-03
88f7356	jens-daniel-mueller	2021-08-02
d759279	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c9ccc00	jens-daniel-mueller	2021-07-22
910d64a	jens-daniel-mueller	2021-07-02
1cbf907	jens-daniel-mueller	2021-07-02
9480f94	jens-daniel-mueller	2021-06-29
7e1f407	jens-daniel-mueller	2021-06-10
2cbe18c	jens-daniel-mueller	2021-06-10
594ed9a	jens-daniel-mueller	2021-06-04
db7df0e	jens-daniel-mueller	2021-06-04
207339d	jens-daniel-mueller	2021-06-03
315710b	jens-daniel-mueller	2021-06-03
d37a85d	jens-daniel-mueller	2021-05-31
25bd183	jens-daniel-mueller	2021-05-26
62bd574	jens-daniel-mueller	2021-05-20
7c56c39	jens-daniel-mueller	2021-05-19
52e7583	jens-daniel-mueller	2021-05-12
969e631	jens-daniel-mueller	2021-05-12
d2a83bc	jens-daniel-mueller	2021-04-16
c0a47df	jens-daniel-mueller	2021-04-16
50290e8	jens-daniel-mueller	2021-04-16
b6fe355	jens-daniel-mueller	2021-04-16
ddec5b7	jens-daniel-mueller	2021-04-15
29edae5	jens-daniel-mueller	2021-04-14
9f31fe3	jens-daniel-mueller	2021-04-13
338dd3c	jens-daniel-mueller	2021-04-09
a79ca2c	jens-daniel-mueller	2021-04-09
eb827c9	jens-daniel-mueller	2021-04-07
857bad3	jens-daniel-mueller	2021-03-24
03b6009	jens-daniel-mueller	2021-03-23
555750f	jens-daniel-mueller	2021-03-23
f155edd	jens-daniel-mueller	2021-03-23
a1d52ff	jens-daniel-mueller	2021-03-15
0bade3b	jens-daniel-mueller	2021-03-15
27c1f4b	jens-daniel-mueller	2021-03-14
af75ebf	jens-daniel-mueller	2021-03-14
5017709	jens-daniel-mueller	2021-03-11
85a5ed2	jens-daniel-mueller	2021-03-10
af70b94	jens-daniel-mueller	2021-03-04
7b672f7	jens-daniel-mueller	2021-01-11
33ba23c	jens-daniel-mueller	2021-01-07
318609d	jens-daniel-mueller	2020-12-23
9d0b2d0	jens-daniel-mueller	2020-12-23
0aa2b50	jens-daniel-mueller	2020-12-23
39113c3	jens-daniel-mueller	2020-12-23
2886da0	jens-daniel-mueller	2020-12-19
02f0ee9	jens-daniel-mueller	2020-12-18
5d452fe	jens-daniel-mueller	2020-12-18
7bcb4eb	jens-daniel-mueller	2020-12-18

tcant_3d_obs <- tcant_3d_obs %>% 
  select(-n_group)

rm(GLODAP_obs_grid_era)

GLODAP_tcant_obs <- full_join(GLODAP, tcant_3d_obs)

rm(GLODAP, tcant_3d_obs)

# fill number of cant data points per grid cell to all observations
GLODAP_tcant_obs <- GLODAP_tcant_obs %>%
  group_by(lon, lat, era, data_source) %>% 
  fill(n, .direction = "updown") %>% 
  ungroup()

The mapped Cant product was merged with GLODAP observation by:

using an identical 1x1° horizontal grid
linear interpolation of C_ant from standard to sampling depth

# interpolate cant to observation depth
GLODAP_tcant_obs_int <- GLODAP_tcant_obs %>%
  filter(n > 1) %>% 
  group_by(lat, lon, era, data_source) %>%
  arrange(depth) %>%
  mutate(tcant_pos_int = approxfun(depth, tcant_pos, rule = 2)(depth)) %>%
  ungroup()

# set cant for observation depth if only one cant available
GLODAP_tcant_obs_set <- GLODAP_tcant_obs %>%
  filter(n == 1) %>%
  group_by(lat, lon, era, data_source) %>%
  mutate(tcant_pos_int = mean(tcant_pos, na.rm = TRUE)) %>%
  ungroup()

# bin data sets with interpolated and set cant
GLODAP_tcant_obs <- bind_rows(GLODAP_tcant_obs_int, GLODAP_tcant_obs_set)
rm(GLODAP_tcant_obs_int, GLODAP_tcant_obs_set)


# remove cant data at grid cells without observations
GLODAP <- GLODAP_tcant_obs %>%
  filter(!is.na(cstar)) %>%
  mutate(tcant_pos = tcant_pos_int) %>%
  select(-tcant_pos_int, n)

rm(GLODAP_tcant_obs)

4.3.7 Merge GLODAP + atm. pCO₂

GLODAP observations were merged with mean annual atmospheric pCO₂ levels by year.

GLODAP <- left_join(GLODAP, co2_atm)

4.3.8 Calculation

# assign reference year
GLODAP <- full_join(GLODAP, tref)

# extract atm pCO2 at reference year
co2_atm_tref <- right_join(co2_atm, tref %>% rename(year = median_year)) %>% 
  select(-year) %>% 
  rename(pCO2_tref = pCO2)

# merge atm pCO2 at tref with GLODAP
GLODAP <- full_join(GLODAP, co2_atm_tref)
rm(co2_atm)


# calculate cstar for reference year
GLODAP <- GLODAP %>%
  mutate(
    cstar_tref_delta =
      ((pCO2 - pCO2_tref) / (pCO2_tref - params_local$preind_atm_pCO2)) * tcant_pos,
    cstar_tref = cstar - cstar_tref_delta)

4.4 Control plots

GLODAP %>% 
  ggplot(aes(cstar_tref_delta)) +
  geom_histogram(binwidth = 1) +
  labs(title = "Histogramm with binwidth = 1") +
  facet_wrap(~ data_source)

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
29444a1	jens-daniel-mueller	2021-08-05
42e80c0	jens-daniel-mueller	2021-08-04
48f6eed	jens-daniel-mueller	2021-08-04
81a46a4	jens-daniel-mueller	2021-08-03
b88c61b	jens-daniel-mueller	2021-08-03
88f7356	jens-daniel-mueller	2021-08-02
d759279	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c9ccc00	jens-daniel-mueller	2021-07-22
c18a8b1	jens-daniel-mueller	2021-07-09
910d64a	jens-daniel-mueller	2021-07-02
1cbf907	jens-daniel-mueller	2021-07-02
3b6387f	jens-daniel-mueller	2021-06-18
c6b3da6	jens-daniel-mueller	2021-06-14
439ee80	jens-daniel-mueller	2021-06-11
33ffcab	jens-daniel-mueller	2021-06-10
7e1f407	jens-daniel-mueller	2021-06-10
2cbe18c	jens-daniel-mueller	2021-06-10
c5aaa55	jens-daniel-mueller	2021-06-10
69c79d0	jens-daniel-mueller	2021-06-08
1772903	jens-daniel-mueller	2021-06-07
594ed9a	jens-daniel-mueller	2021-06-04
db7df0e	jens-daniel-mueller	2021-06-04
2edc791	jens-daniel-mueller	2021-06-04
207339d	jens-daniel-mueller	2021-06-03
315710b	jens-daniel-mueller	2021-06-03
be90356	jens-daniel-mueller	2021-06-02
d37a85d	jens-daniel-mueller	2021-05-31
25bd183	jens-daniel-mueller	2021-05-26
62bd574	jens-daniel-mueller	2021-05-20
7c56c39	jens-daniel-mueller	2021-05-19
52e7583	jens-daniel-mueller	2021-05-12
969e631	jens-daniel-mueller	2021-05-12
d2a83bc	jens-daniel-mueller	2021-04-16
c0a47df	jens-daniel-mueller	2021-04-16
50290e8	jens-daniel-mueller	2021-04-16
a00ec94	jens-daniel-mueller	2021-04-16
b6fe355	jens-daniel-mueller	2021-04-16
ddec5b7	jens-daniel-mueller	2021-04-15
29edae5	jens-daniel-mueller	2021-04-14
9f31fe3	jens-daniel-mueller	2021-04-13
338dd3c	jens-daniel-mueller	2021-04-09
a79ca2c	jens-daniel-mueller	2021-04-09
be095c6	jens-daniel-mueller	2021-04-09
eb827c9	jens-daniel-mueller	2021-04-07
857bad3	jens-daniel-mueller	2021-03-24
03b6009	jens-daniel-mueller	2021-03-23
555750f	jens-daniel-mueller	2021-03-23
a1d52ff	jens-daniel-mueller	2021-03-15
0bade3b	jens-daniel-mueller	2021-03-15
27c1f4b	jens-daniel-mueller	2021-03-14
af75ebf	jens-daniel-mueller	2021-03-14
5017709	jens-daniel-mueller	2021-03-11
85a5ed2	jens-daniel-mueller	2021-03-10
3c2ec33	jens-daniel-mueller	2021-03-05
af70b94	jens-daniel-mueller	2021-03-04
7b672f7	jens-daniel-mueller	2021-01-11
33ba23c	jens-daniel-mueller	2021-01-07
318609d	jens-daniel-mueller	2020-12-23
9d0b2d0	jens-daniel-mueller	2020-12-23
0aa2b50	jens-daniel-mueller	2020-12-23
2886da0	jens-daniel-mueller	2020-12-19
02f0ee9	jens-daniel-mueller	2020-12-18
7bcb4eb	jens-daniel-mueller	2020-12-18
158fe26	jens-daniel-mueller	2020-12-15
3ebff89	jens-daniel-mueller	2020-12-12
7d82772	jens-daniel-mueller	2020-12-11
7788175	jens-daniel-mueller	2020-12-09
24a632f	jens-daniel-mueller	2020-12-07
6a8004b	jens-daniel-mueller	2020-12-07
70bf1a5	jens-daniel-mueller	2020-12-07
7555355	jens-daniel-mueller	2020-12-07
143d6fa	jens-daniel-mueller	2020-12-07
37e9dac	jens-daniel-mueller	2020-12-02
0ff728b	jens-daniel-mueller	2020-12-01
b02b7a4	jens-daniel-mueller	2020-12-01
196be51	jens-daniel-mueller	2020-11-30
bc61ce3	Jens Müller	2020-11-30

GLODAP %>% 
  sample_n(1e4) %>% 
  ggplot(aes(year, cstar_tref_delta, col = tcant_pos)) +
  geom_point() +
  scale_color_viridis_c() +
  labs(title = "Time series of random subsample 1e4") +
  facet_wrap(~ data_source)

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
29444a1	jens-daniel-mueller	2021-08-05
42e80c0	jens-daniel-mueller	2021-08-04
48f6eed	jens-daniel-mueller	2021-08-04
81a46a4	jens-daniel-mueller	2021-08-03
b88c61b	jens-daniel-mueller	2021-08-03
88f7356	jens-daniel-mueller	2021-08-02
d759279	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c9ccc00	jens-daniel-mueller	2021-07-22
c18a8b1	jens-daniel-mueller	2021-07-09
910d64a	jens-daniel-mueller	2021-07-02
1cbf907	jens-daniel-mueller	2021-07-02
9480f94	jens-daniel-mueller	2021-06-29
3b6387f	jens-daniel-mueller	2021-06-18
77cfc43	jens-daniel-mueller	2021-06-17
c6b3da6	jens-daniel-mueller	2021-06-14
439ee80	jens-daniel-mueller	2021-06-11
33ffcab	jens-daniel-mueller	2021-06-10
7e1f407	jens-daniel-mueller	2021-06-10
2cbe18c	jens-daniel-mueller	2021-06-10
c5aaa55	jens-daniel-mueller	2021-06-10
69c79d0	jens-daniel-mueller	2021-06-08
1772903	jens-daniel-mueller	2021-06-07
594ed9a	jens-daniel-mueller	2021-06-04
db7df0e	jens-daniel-mueller	2021-06-04
2edc791	jens-daniel-mueller	2021-06-04
207339d	jens-daniel-mueller	2021-06-03
315710b	jens-daniel-mueller	2021-06-03
be90356	jens-daniel-mueller	2021-06-02
d37a85d	jens-daniel-mueller	2021-05-31
25bd183	jens-daniel-mueller	2021-05-26
62bd574	jens-daniel-mueller	2021-05-20
7c56c39	jens-daniel-mueller	2021-05-19
52e7583	jens-daniel-mueller	2021-05-12
969e631	jens-daniel-mueller	2021-05-12
d2a83bc	jens-daniel-mueller	2021-04-16
c0a47df	jens-daniel-mueller	2021-04-16
50290e8	jens-daniel-mueller	2021-04-16
a00ec94	jens-daniel-mueller	2021-04-16
b6fe355	jens-daniel-mueller	2021-04-16
ddec5b7	jens-daniel-mueller	2021-04-15
29edae5	jens-daniel-mueller	2021-04-14
9f31fe3	jens-daniel-mueller	2021-04-13
338dd3c	jens-daniel-mueller	2021-04-09
a79ca2c	jens-daniel-mueller	2021-04-09
be095c6	jens-daniel-mueller	2021-04-09
eb827c9	jens-daniel-mueller	2021-04-07
857bad3	jens-daniel-mueller	2021-03-24
03b6009	jens-daniel-mueller	2021-03-23
555750f	jens-daniel-mueller	2021-03-23
a1d52ff	jens-daniel-mueller	2021-03-15
0bade3b	jens-daniel-mueller	2021-03-15
27c1f4b	jens-daniel-mueller	2021-03-14
af75ebf	jens-daniel-mueller	2021-03-14
5017709	jens-daniel-mueller	2021-03-11
85a5ed2	jens-daniel-mueller	2021-03-10
af70b94	jens-daniel-mueller	2021-03-04
7b672f7	jens-daniel-mueller	2021-01-11
33ba23c	jens-daniel-mueller	2021-01-07
318609d	jens-daniel-mueller	2020-12-23
9d0b2d0	jens-daniel-mueller	2020-12-23
0aa2b50	jens-daniel-mueller	2020-12-23
2886da0	jens-daniel-mueller	2020-12-19
02f0ee9	jens-daniel-mueller	2020-12-18
5d452fe	jens-daniel-mueller	2020-12-18
7bcb4eb	jens-daniel-mueller	2020-12-18
158fe26	jens-daniel-mueller	2020-12-15
3ebff89	jens-daniel-mueller	2020-12-12
7d82772	jens-daniel-mueller	2020-12-11
7788175	jens-daniel-mueller	2020-12-09
24a632f	jens-daniel-mueller	2020-12-07
6a8004b	jens-daniel-mueller	2020-12-07
70bf1a5	jens-daniel-mueller	2020-12-07
7555355	jens-daniel-mueller	2020-12-07
143d6fa	jens-daniel-mueller	2020-12-07
37e9dac	jens-daniel-mueller	2020-12-02
0ff728b	jens-daniel-mueller	2020-12-01
b02b7a4	jens-daniel-mueller	2020-12-01
196be51	jens-daniel-mueller	2020-11-30
bc61ce3	Jens Müller	2020-11-30

GLODAP %>% 
  ggplot(aes(year, cstar_tref_delta)) +
  geom_bin2d(binwidth = 1) +
  scale_fill_viridis_c(trans = "log10") +
  labs(title = "Heatmap with binwidth = 1") +
  facet_wrap(~ data_source)

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
29444a1	jens-daniel-mueller	2021-08-05
42e80c0	jens-daniel-mueller	2021-08-04
48f6eed	jens-daniel-mueller	2021-08-04
81a46a4	jens-daniel-mueller	2021-08-03
b88c61b	jens-daniel-mueller	2021-08-03
88f7356	jens-daniel-mueller	2021-08-02
d759279	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c9ccc00	jens-daniel-mueller	2021-07-22
c18a8b1	jens-daniel-mueller	2021-07-09
910d64a	jens-daniel-mueller	2021-07-02
1cbf907	jens-daniel-mueller	2021-07-02
9480f94	jens-daniel-mueller	2021-06-29
3b6387f	jens-daniel-mueller	2021-06-18
77cfc43	jens-daniel-mueller	2021-06-17
c6b3da6	jens-daniel-mueller	2021-06-14
439ee80	jens-daniel-mueller	2021-06-11
33ffcab	jens-daniel-mueller	2021-06-10
7e1f407	jens-daniel-mueller	2021-06-10
2cbe18c	jens-daniel-mueller	2021-06-10
c5aaa55	jens-daniel-mueller	2021-06-10
69c79d0	jens-daniel-mueller	2021-06-08
1772903	jens-daniel-mueller	2021-06-07
594ed9a	jens-daniel-mueller	2021-06-04
db7df0e	jens-daniel-mueller	2021-06-04
2edc791	jens-daniel-mueller	2021-06-04
207339d	jens-daniel-mueller	2021-06-03
315710b	jens-daniel-mueller	2021-06-03
be90356	jens-daniel-mueller	2021-06-02
d37a85d	jens-daniel-mueller	2021-05-31
25bd183	jens-daniel-mueller	2021-05-26
62bd574	jens-daniel-mueller	2021-05-20
7c56c39	jens-daniel-mueller	2021-05-19
52e7583	jens-daniel-mueller	2021-05-12
969e631	jens-daniel-mueller	2021-05-12
d2a83bc	jens-daniel-mueller	2021-04-16
c0a47df	jens-daniel-mueller	2021-04-16
50290e8	jens-daniel-mueller	2021-04-16
a00ec94	jens-daniel-mueller	2021-04-16
b6fe355	jens-daniel-mueller	2021-04-16
ddec5b7	jens-daniel-mueller	2021-04-15
29edae5	jens-daniel-mueller	2021-04-14
9f31fe3	jens-daniel-mueller	2021-04-13
338dd3c	jens-daniel-mueller	2021-04-09
a79ca2c	jens-daniel-mueller	2021-04-09
be095c6	jens-daniel-mueller	2021-04-09
eb827c9	jens-daniel-mueller	2021-04-07
857bad3	jens-daniel-mueller	2021-03-24
03b6009	jens-daniel-mueller	2021-03-23
555750f	jens-daniel-mueller	2021-03-23
a1d52ff	jens-daniel-mueller	2021-03-15
0bade3b	jens-daniel-mueller	2021-03-15
27c1f4b	jens-daniel-mueller	2021-03-14
af75ebf	jens-daniel-mueller	2021-03-14
5017709	jens-daniel-mueller	2021-03-11
85a5ed2	jens-daniel-mueller	2021-03-10
3c2ec33	jens-daniel-mueller	2021-03-05
af70b94	jens-daniel-mueller	2021-03-04
7b672f7	jens-daniel-mueller	2021-01-11
33ba23c	jens-daniel-mueller	2021-01-07
318609d	jens-daniel-mueller	2020-12-23
9d0b2d0	jens-daniel-mueller	2020-12-23
0aa2b50	jens-daniel-mueller	2020-12-23
2886da0	jens-daniel-mueller	2020-12-19
02f0ee9	jens-daniel-mueller	2020-12-18
7bcb4eb	jens-daniel-mueller	2020-12-18
158fe26	jens-daniel-mueller	2020-12-15
3ebff89	jens-daniel-mueller	2020-12-12
7d82772	jens-daniel-mueller	2020-12-11
7788175	jens-daniel-mueller	2020-12-09
24a632f	jens-daniel-mueller	2020-12-07
6a8004b	jens-daniel-mueller	2020-12-07
70bf1a5	jens-daniel-mueller	2020-12-07
7555355	jens-daniel-mueller	2020-12-07
143d6fa	jens-daniel-mueller	2020-12-07
090e4d5	jens-daniel-mueller	2020-12-02
37e9dac	jens-daniel-mueller	2020-12-02
0ff728b	jens-daniel-mueller	2020-12-01
b02b7a4	jens-daniel-mueller	2020-12-01
196be51	jens-daniel-mueller	2020-11-30
bc61ce3	Jens Müller	2020-11-30

5 Isoneutral slabs

GLODAP %>% 
  ggplot(aes(gamma)) +
  geom_histogram() +
  facet_grid(basin_AIP~data_source)

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
29444a1	jens-daniel-mueller	2021-08-05
42e80c0	jens-daniel-mueller	2021-08-04
48f6eed	jens-daniel-mueller	2021-08-04
81a46a4	jens-daniel-mueller	2021-08-03
b88c61b	jens-daniel-mueller	2021-08-03
88f7356	jens-daniel-mueller	2021-08-02
d759279	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c9ccc00	jens-daniel-mueller	2021-07-22
910d64a	jens-daniel-mueller	2021-07-02
1cbf907	jens-daniel-mueller	2021-07-02
7e1f407	jens-daniel-mueller	2021-06-10
2cbe18c	jens-daniel-mueller	2021-06-10
594ed9a	jens-daniel-mueller	2021-06-04
db7df0e	jens-daniel-mueller	2021-06-04
207339d	jens-daniel-mueller	2021-06-03
315710b	jens-daniel-mueller	2021-06-03
d37a85d	jens-daniel-mueller	2021-05-31
25bd183	jens-daniel-mueller	2021-05-26
62bd574	jens-daniel-mueller	2021-05-20
7c56c39	jens-daniel-mueller	2021-05-19

GLODAP %>% 
  ggplot(aes(gamma, depth)) +
  geom_bin2d() +
  scale_y_reverse() +
  scale_fill_viridis_c() +
  facet_grid(basin_AIP~data_source)

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
29444a1	jens-daniel-mueller	2021-08-05
42e80c0	jens-daniel-mueller	2021-08-04
48f6eed	jens-daniel-mueller	2021-08-04
81a46a4	jens-daniel-mueller	2021-08-03
b88c61b	jens-daniel-mueller	2021-08-03
88f7356	jens-daniel-mueller	2021-08-02
d759279	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c9ccc00	jens-daniel-mueller	2021-07-22
c18a8b1	jens-daniel-mueller	2021-07-09
910d64a	jens-daniel-mueller	2021-07-02
1cbf907	jens-daniel-mueller	2021-07-02
9480f94	jens-daniel-mueller	2021-06-29
7e1f407	jens-daniel-mueller	2021-06-10
2cbe18c	jens-daniel-mueller	2021-06-10
594ed9a	jens-daniel-mueller	2021-06-04
db7df0e	jens-daniel-mueller	2021-06-04
207339d	jens-daniel-mueller	2021-06-03
315710b	jens-daniel-mueller	2021-06-03
d37a85d	jens-daniel-mueller	2021-05-31
25bd183	jens-daniel-mueller	2021-05-26
62bd574	jens-daniel-mueller	2021-05-20
7c56c39	jens-daniel-mueller	2021-05-19

GLODAP <- GLODAP %>% 
  group_by(basin_AIP) %>% 
  mutate(gamma_slab = cut_number(gamma, n = 10)) %>% 
  ungroup()

GLODAP_slab <- GLODAP %>% 
  count(gamma_slab, basin_AIP, era)


GLODAP_slab %>% 
  ggplot(aes(gamma_slab, n, fill=era)) +
  geom_col() +
  facet_grid(.~basin_AIP, scales = "free_x") +
  theme(axis.text.x = element_text(angle = 90))

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
29444a1	jens-daniel-mueller	2021-08-05
42e80c0	jens-daniel-mueller	2021-08-04
48f6eed	jens-daniel-mueller	2021-08-04
81a46a4	jens-daniel-mueller	2021-08-03
b88c61b	jens-daniel-mueller	2021-08-03
88f7356	jens-daniel-mueller	2021-08-02
d759279	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c9ccc00	jens-daniel-mueller	2021-07-22
910d64a	jens-daniel-mueller	2021-07-02
1cbf907	jens-daniel-mueller	2021-07-02
9480f94	jens-daniel-mueller	2021-06-29
7e1f407	jens-daniel-mueller	2021-06-10
2cbe18c	jens-daniel-mueller	2021-06-10
594ed9a	jens-daniel-mueller	2021-06-04
db7df0e	jens-daniel-mueller	2021-06-04
207339d	jens-daniel-mueller	2021-06-03
315710b	jens-daniel-mueller	2021-06-03
d37a85d	jens-daniel-mueller	2021-05-31
25bd183	jens-daniel-mueller	2021-05-26
62bd574	jens-daniel-mueller	2021-05-20
7c56c39	jens-daniel-mueller	2021-05-19

The following boundaries for isoneutral slabs were defined:

Atlantic: -, 26, 26.5, 26.75, 27, 27.25, 27.5, 27.75, 27.85, 27.95, 28.05, 28.1, 28.15, 28.2,
Indo-Pacific: -, 26, 26.5, 26.75, 27, 27.25, 27.5, 27.75, 27.85, 27.95, 28.05, 28.1, 28.2, 28.3, 28.4,

Continuous neutral densities (gamma) values from GLODAP are grouped into isoneutral slabs.

GLODAP <- m_cut_gamma(GLODAP, "gamma")

# this section was only used to calculate gamma locally, and compare it to the value provided in GLODAP data set

GLODAP_cruise <- GLODAP %>% 
  filter(cruise %in% params_global$cruises_meridional)

library(oce)
library(gsw)
# calculate pressure from depth

GLODAP_cruise <- GLODAP_cruise %>% 
  mutate(CTDPRS = gsw_p_from_z(-depth,
                               lat))

GLODAP_cruise <- GLODAP_cruise %>% 
  mutate(THETA = swTheta(salinity = sal,
                         temperature = temp,
                         pressure = CTDPRS,
                         referencePressure = 0,
                         longitude = lon-180,
                         latitude = lat))

GLODAP_cruise <- GLODAP_cruise %>% 
  rename(LATITUDE = lat,
         LONGITUDE = lon,
         SALNTY = sal,
         gamma_provided = gamma)

library(reticulate)
data_source_python(here::here("code/python_scripts",
                         "Gamma_GLODAP_python.py"))

GLODAP_cruise <- calculate_gamma(GLODAP_cruise)

GLODAP_cruise <- GLODAP_cruise %>% 
  mutate(gamma_delta = gamma_provided - GAMMA)

lat_section <- 
GLODAP_cruise %>%
  ggplot(aes(LATITUDE, CTDPRS)) +
  scale_y_reverse() +
  theme(legend.position = "bottom")

lat_section +
  stat_summary_2d(aes(z = gamma_delta)) +
  scale_color_viridis_c()

GLODAP_cruise %>% 
  ggplot(aes(gamma_delta))+
  geom_histogram()

rm(lat_section, GLODAP_cruise, cruises_meridional)

6 Rarefication

# apply coarse grid
GLODAP <- m_grid_horizontal_coarse(GLODAP)

# count data in each coarse grid element
GLODAP_grid <- GLODAP %>% 
  group_by(lat_grid, lon_grid, gamma_slab, era, data_source) %>% 
  count() %>% 
  ungroup()

GLODAP_grid %>%
  ggplot(aes(n)) +
  geom_histogram(binwidth = 1) +
  facet_grid(data_source ~ .) +
  labs(title = "Observations per era, grid cell and slab, all data")

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
da61d1a	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
29444a1	jens-daniel-mueller	2021-08-05
42e80c0	jens-daniel-mueller	2021-08-04
48f6eed	jens-daniel-mueller	2021-08-04
81a46a4	jens-daniel-mueller	2021-08-03
b88c61b	jens-daniel-mueller	2021-08-03
88f7356	jens-daniel-mueller	2021-08-02
d759279	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c9ccc00	jens-daniel-mueller	2021-07-22
c18a8b1	jens-daniel-mueller	2021-07-09
910d64a	jens-daniel-mueller	2021-07-02
1cbf907	jens-daniel-mueller	2021-07-02
6f4c8e3	jens-daniel-mueller	2021-06-30
749d84e	jens-daniel-mueller	2021-06-30

GLODAP_grid %>%
  filter(n <= quantile(n, 0.9)) %>%
  ggplot(aes(n)) +
  geom_histogram(binwidth = 1) +
  facet_grid(data_source ~ .) +
  labs(title = "Observations per era, grid cell and slab, 90th quantile")

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
da61d1a	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
29444a1	jens-daniel-mueller	2021-08-05
42e80c0	jens-daniel-mueller	2021-08-04
48f6eed	jens-daniel-mueller	2021-08-04
81a46a4	jens-daniel-mueller	2021-08-03
b88c61b	jens-daniel-mueller	2021-08-03
88f7356	jens-daniel-mueller	2021-08-02
d759279	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c9ccc00	jens-daniel-mueller	2021-07-22
c18a8b1	jens-daniel-mueller	2021-07-09

GLODAP_grid %>%
  group_by(gamma_slab) %>%
  group_split() %>%
  # head(1) %>% 
  map(
    ~ map +
      geom_raster(data = .x,
                  aes(lon_grid, lat_grid, fill = n)) +
      facet_grid(data_source ~ era) +
      scale_fill_viridis_c(trans = "log10") +
      labs(title = paste("Before rarefication | gamma_slab",
                         unique(.x$gamma_slab)))
  )

[[1]]

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
da61d1a	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
29444a1	jens-daniel-mueller	2021-08-05
42e80c0	jens-daniel-mueller	2021-08-04
48f6eed	jens-daniel-mueller	2021-08-04
81a46a4	jens-daniel-mueller	2021-08-03
b88c61b	jens-daniel-mueller	2021-08-03
88f7356	jens-daniel-mueller	2021-08-02
d759279	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c9ccc00	jens-daniel-mueller	2021-07-22
910d64a	jens-daniel-mueller	2021-07-02
1cbf907	jens-daniel-mueller	2021-07-02
6f4c8e3	jens-daniel-mueller	2021-06-30
955f5dc	jens-daniel-mueller	2021-06-29
9480f94	jens-daniel-mueller	2021-06-29
d054c39	jens-daniel-mueller	2021-06-24
87e2739	jens-daniel-mueller	2021-06-24


[[2]]

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
da61d1a	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
29444a1	jens-daniel-mueller	2021-08-05
42e80c0	jens-daniel-mueller	2021-08-04
48f6eed	jens-daniel-mueller	2021-08-04
81a46a4	jens-daniel-mueller	2021-08-03
b88c61b	jens-daniel-mueller	2021-08-03
88f7356	jens-daniel-mueller	2021-08-02
d759279	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c9ccc00	jens-daniel-mueller	2021-07-22
910d64a	jens-daniel-mueller	2021-07-02
1cbf907	jens-daniel-mueller	2021-07-02
6f4c8e3	jens-daniel-mueller	2021-06-30
955f5dc	jens-daniel-mueller	2021-06-29
9480f94	jens-daniel-mueller	2021-06-29
d054c39	jens-daniel-mueller	2021-06-24
87e2739	jens-daniel-mueller	2021-06-24


[[3]]


[[4]]


[[5]]


[[6]]


[[7]]


[[8]]


[[9]]


[[10]]


[[11]]


[[12]]


[[13]]


[[14]]


[[15]]

if (params_local$rarefication == "coarse_grid") {

GLODAP <- GLODAP %>% 
  group_by(lat_grid, lon_grid, gamma_slab, era, data_source) %>% 
  slice_sample(n = params_local$rarefication_threshold) %>% 
  ungroup()

} else {print("No rarefication applied")}


GLODAP_grid <- GLODAP %>% 
  group_by(lat_grid, lon_grid, gamma_slab, era, data_source) %>% 
  count() %>% 
  ungroup()

GLODAP_grid %>%
  group_by(gamma_slab) %>%
  group_split() %>%
  # head(1) %>%
  map(
    ~ map +
      geom_raster(data = .x,
                  aes(lon_grid, lat_grid, fill = n)) +
      facet_grid(data_source ~ era) +
      scale_fill_viridis_c(trans = "log10") +
      labs(title = paste("After rarefication | gamma_slab",
                         unique(.x$gamma_slab)))
  )

[[1]]


[[2]]


[[3]]


[[4]]


[[5]]


[[6]]


[[7]]


[[8]]


[[9]]


[[10]]


[[11]]


[[12]]


[[13]]


[[14]]


[[15]]

GLODAP <- GLODAP %>% 
  select(-c(lat_grid, lon_grid))

rm(GLODAP_grid)

7 Observations coverage

GLODAP <- GLODAP %>% 
  mutate(gamma_slab = factor(gamma_slab), 
         gamma_slab = factor(gamma_slab, levels = rev(levels(gamma_slab))))

for (i_basin in unique(GLODAP$basin)) {
  # i_basin <- unique(GLODAP$basin)[1]
  
  print(
    GLODAP %>%
      filter(basin == i_basin) %>%
      ggplot(aes(lat, gamma_slab)) +
      geom_bin2d(binwidth = 5) +
      scale_fill_viridis_c(
        option = "magma",
        direction = -1,
        trans = "log10"
      ) +
      scale_x_continuous(breaks = seq(-100, 100, 20),
                         limits = c(params_global$lat_min,
                                    params_global$lat_max)) +
      facet_grid(era ~ data_source) +
      labs(title = paste("MLR region: ", i_basin))
  )
  
}

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
da61d1a	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
29444a1	jens-daniel-mueller	2021-08-05
42e80c0	jens-daniel-mueller	2021-08-04
48f6eed	jens-daniel-mueller	2021-08-04
81a46a4	jens-daniel-mueller	2021-08-03
b88c61b	jens-daniel-mueller	2021-08-03
88f7356	jens-daniel-mueller	2021-08-02
d759279	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c9ccc00	jens-daniel-mueller	2021-07-22
910d64a	jens-daniel-mueller	2021-07-02
1cbf907	jens-daniel-mueller	2021-07-02
955f5dc	jens-daniel-mueller	2021-06-29
9480f94	jens-daniel-mueller	2021-06-29
d054c39	jens-daniel-mueller	2021-06-24
87e2739	jens-daniel-mueller	2021-06-24
4fcd066	jens-daniel-mueller	2021-06-24
7e1f407	jens-daniel-mueller	2021-06-10
2cbe18c	jens-daniel-mueller	2021-06-10
594ed9a	jens-daniel-mueller	2021-06-04
db7df0e	jens-daniel-mueller	2021-06-04
207339d	jens-daniel-mueller	2021-06-03
315710b	jens-daniel-mueller	2021-06-03
d37a85d	jens-daniel-mueller	2021-05-31
25bd183	jens-daniel-mueller	2021-05-26
62bd574	jens-daniel-mueller	2021-05-20
7c56c39	jens-daniel-mueller	2021-05-19
52e7583	jens-daniel-mueller	2021-05-12
d2a83bc	jens-daniel-mueller	2021-04-16
c0a47df	jens-daniel-mueller	2021-04-16
50290e8	jens-daniel-mueller	2021-04-16
b6fe355	jens-daniel-mueller	2021-04-16
81b7c6d	jens-daniel-mueller	2021-04-16
ddec5b7	jens-daniel-mueller	2021-04-15
29edae5	jens-daniel-mueller	2021-04-14
9f31fe3	jens-daniel-mueller	2021-04-13
338dd3c	jens-daniel-mueller	2021-04-09
a79ca2c	jens-daniel-mueller	2021-04-09
eb827c9	jens-daniel-mueller	2021-04-07
857bad3	jens-daniel-mueller	2021-03-24
03b6009	jens-daniel-mueller	2021-03-23
555750f	jens-daniel-mueller	2021-03-23
f155edd	jens-daniel-mueller	2021-03-23
83a13de	jens-daniel-mueller	2021-03-20
cf98c6d	jens-daniel-mueller	2021-03-16
a1d52ff	jens-daniel-mueller	2021-03-15
0bade3b	jens-daniel-mueller	2021-03-15
27c1f4b	jens-daniel-mueller	2021-03-14
5017709	jens-daniel-mueller	2021-03-11
85a5ed2	jens-daniel-mueller	2021-03-10
7b672f7	jens-daniel-mueller	2021-01-11
33ba23c	jens-daniel-mueller	2021-01-07
318609d	jens-daniel-mueller	2020-12-23
9d0b2d0	jens-daniel-mueller	2020-12-23
0aa2b50	jens-daniel-mueller	2020-12-23
2886da0	jens-daniel-mueller	2020-12-19
02f0ee9	jens-daniel-mueller	2020-12-18
158fe26	jens-daniel-mueller	2020-12-15
984697e	jens-daniel-mueller	2020-12-12
3ebff89	jens-daniel-mueller	2020-12-12
7d82772	jens-daniel-mueller	2020-12-11

7.1 Histograms

GLODAP_vars <- GLODAP %>% 
  select(data_source,
         params_local$MLR_target,
         params_local$MLR_predictors)

GLODAP_vars_long <- GLODAP_vars %>%
  pivot_longer(
    cols = c(params_local$MLR_target,
             params_local$MLR_predictors),
    names_to = "variable",
    values_to = "value"
  )

GLODAP_vars_long %>% 
  ggplot(aes(value)) +
  geom_histogram() +
  facet_grid(data_source ~ variable,
             scales = "free_x")

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
da61d1a	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
29444a1	jens-daniel-mueller	2021-08-05
42e80c0	jens-daniel-mueller	2021-08-04
48f6eed	jens-daniel-mueller	2021-08-04
81a46a4	jens-daniel-mueller	2021-08-03
b88c61b	jens-daniel-mueller	2021-08-03
a53656d	jens-daniel-mueller	2021-08-03
88f7356	jens-daniel-mueller	2021-08-02
d759279	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c9ccc00	jens-daniel-mueller	2021-07-22
c18a8b1	jens-daniel-mueller	2021-07-09
910d64a	jens-daniel-mueller	2021-07-02
1cbf907	jens-daniel-mueller	2021-07-02
955f5dc	jens-daniel-mueller	2021-06-29
d054c39	jens-daniel-mueller	2021-06-24
87e2739	jens-daniel-mueller	2021-06-24
4fcd066	jens-daniel-mueller	2021-06-24
3b6387f	jens-daniel-mueller	2021-06-18
c6b3da6	jens-daniel-mueller	2021-06-14
439ee80	jens-daniel-mueller	2021-06-11
33ffcab	jens-daniel-mueller	2021-06-10
7e1f407	jens-daniel-mueller	2021-06-10
2cbe18c	jens-daniel-mueller	2021-06-10
c5aaa55	jens-daniel-mueller	2021-06-10
69c79d0	jens-daniel-mueller	2021-06-08
1772903	jens-daniel-mueller	2021-06-07
594ed9a	jens-daniel-mueller	2021-06-04
db7df0e	jens-daniel-mueller	2021-06-04
2edc791	jens-daniel-mueller	2021-06-04
207339d	jens-daniel-mueller	2021-06-03
315710b	jens-daniel-mueller	2021-06-03
be90356	jens-daniel-mueller	2021-06-02
d37a85d	jens-daniel-mueller	2021-05-31
4b7a5ee	jens-daniel-mueller	2021-05-28
25bd183	jens-daniel-mueller	2021-05-26
62bd574	jens-daniel-mueller	2021-05-20
7c56c39	jens-daniel-mueller	2021-05-19
52e7583	jens-daniel-mueller	2021-05-12
969e631	jens-daniel-mueller	2021-05-12
d2a83bc	jens-daniel-mueller	2021-04-16
c0a47df	jens-daniel-mueller	2021-04-16
50290e8	jens-daniel-mueller	2021-04-16
a00ec94	jens-daniel-mueller	2021-04-16
b6fe355	jens-daniel-mueller	2021-04-16
ddec5b7	jens-daniel-mueller	2021-04-15
29edae5	jens-daniel-mueller	2021-04-14
9f31fe3	jens-daniel-mueller	2021-04-13
338dd3c	jens-daniel-mueller	2021-04-09
a79ca2c	jens-daniel-mueller	2021-04-09
be095c6	jens-daniel-mueller	2021-04-09
eb827c9	jens-daniel-mueller	2021-04-07
857bad3	jens-daniel-mueller	2021-03-24
03b6009	jens-daniel-mueller	2021-03-23
555750f	jens-daniel-mueller	2021-03-23
a1d52ff	jens-daniel-mueller	2021-03-15
0bade3b	jens-daniel-mueller	2021-03-15
27c1f4b	jens-daniel-mueller	2021-03-14
af75ebf	jens-daniel-mueller	2021-03-14
5017709	jens-daniel-mueller	2021-03-11
85a5ed2	jens-daniel-mueller	2021-03-10
af70b94	jens-daniel-mueller	2021-03-04
7b672f7	jens-daniel-mueller	2021-01-11
33ba23c	jens-daniel-mueller	2021-01-07
318609d	jens-daniel-mueller	2020-12-23
9d0b2d0	jens-daniel-mueller	2020-12-23
0aa2b50	jens-daniel-mueller	2020-12-23
2886da0	jens-daniel-mueller	2020-12-19
02f0ee9	jens-daniel-mueller	2020-12-18
7bcb4eb	jens-daniel-mueller	2020-12-18
158fe26	jens-daniel-mueller	2020-12-15
3ebff89	jens-daniel-mueller	2020-12-12
7d82772	jens-daniel-mueller	2020-12-11
7788175	jens-daniel-mueller	2020-12-09

rm(GLODAP_vars, GLODAP_vars_long)

8 Cruise quality check

8.1 MLR models

The the following full MLR model was fitted to all GLODAP, irrespective of the sampling era:

#define full model
model <- paste("cstar",
               paste(params_local$MLR_predictors, collapse = " + "),
               sep = " ~ ")

model

[1] "cstar ~ sal + temp + aou + oxygen + nitrate + silicate"

# prepare nested data frame
GLODAP_nested <- GLODAP %>%
  group_by(gamma_slab, data_source) %>%
  nest()

# expand with model definitions
GLODAP_nested_lm <- expand_grid(GLODAP_nested,
                                model)

# fit models and extract tidy model output
GLODAP_nested_lm_fit <- GLODAP_nested_lm %>%
  mutate(
    fit = map2(.x = data, .y = model,
               ~ lm(as.formula(.y), data = .x)),
    tidied = map(fit, tidy),
    glanced = map(fit, glance),
    augmented = map(fit, augment)
  )

# extract augmented model output (fitted values and residuals)
GLODAP_augmented <- GLODAP_nested_lm_fit %>%
  select(-c(data, fit, tidied, glanced)) %>%
  unnest(augmented)

# extract input data
GLODAP_data <- GLODAP_nested_lm_fit %>%
  select(-c(fit, tidied, glanced, augmented)) %>%
  unnest(data)

# append input data with augmented data
GLODAP_augmented <- bind_cols(GLODAP_data,
                              GLODAP_augmented %>% select(.fitted, .resid))

rm(GLODAP_data)

8.1.1 Residuals

Below, the residuals of C* from the mean C* and from C* predicted with the global model are shown.

GLODAP_augmented <- GLODAP_augmented %>%
  group_by(data_source) %>% 
  mutate(cstar_minus_mean = cstar_tref - mean(cstar_tref)) %>% 
  ungroup() 

GLODAP_augmented%>% 
  ggplot(aes(year, cstar_minus_mean)) +
  geom_hline(yintercept = 0) +
  geom_bin2d(binwidth = c(1, 1)) +
  scale_fill_viridis_c() +
  facet_grid(basin_AIP ~ data_source)

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
da61d1a	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
29444a1	jens-daniel-mueller	2021-08-05
42e80c0	jens-daniel-mueller	2021-08-04
48f6eed	jens-daniel-mueller	2021-08-04
81a46a4	jens-daniel-mueller	2021-08-03
b88c61b	jens-daniel-mueller	2021-08-03
88f7356	jens-daniel-mueller	2021-08-02
d759279	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c9ccc00	jens-daniel-mueller	2021-07-22
c18a8b1	jens-daniel-mueller	2021-07-09
910d64a	jens-daniel-mueller	2021-07-02
1cbf907	jens-daniel-mueller	2021-07-02
955f5dc	jens-daniel-mueller	2021-06-29
9480f94	jens-daniel-mueller	2021-06-29
d054c39	jens-daniel-mueller	2021-06-24
87e2739	jens-daniel-mueller	2021-06-24
4fcd066	jens-daniel-mueller	2021-06-24
3b6387f	jens-daniel-mueller	2021-06-18
77cfc43	jens-daniel-mueller	2021-06-17
c6b3da6	jens-daniel-mueller	2021-06-14
439ee80	jens-daniel-mueller	2021-06-11
33ffcab	jens-daniel-mueller	2021-06-10
7e1f407	jens-daniel-mueller	2021-06-10
2cbe18c	jens-daniel-mueller	2021-06-10
c5aaa55	jens-daniel-mueller	2021-06-10
69c79d0	jens-daniel-mueller	2021-06-08
1772903	jens-daniel-mueller	2021-06-07
594ed9a	jens-daniel-mueller	2021-06-04
db7df0e	jens-daniel-mueller	2021-06-04
2edc791	jens-daniel-mueller	2021-06-04
207339d	jens-daniel-mueller	2021-06-03
315710b	jens-daniel-mueller	2021-06-03
be90356	jens-daniel-mueller	2021-06-02
d37a85d	jens-daniel-mueller	2021-05-31
25bd183	jens-daniel-mueller	2021-05-26
62bd574	jens-daniel-mueller	2021-05-20
7c56c39	jens-daniel-mueller	2021-05-19
52e7583	jens-daniel-mueller	2021-05-12
969e631	jens-daniel-mueller	2021-05-12
d2a83bc	jens-daniel-mueller	2021-04-16
c0a47df	jens-daniel-mueller	2021-04-16
50290e8	jens-daniel-mueller	2021-04-16
a00ec94	jens-daniel-mueller	2021-04-16
b6fe355	jens-daniel-mueller	2021-04-16
ddec5b7	jens-daniel-mueller	2021-04-15
29edae5	jens-daniel-mueller	2021-04-14
9f31fe3	jens-daniel-mueller	2021-04-13
338dd3c	jens-daniel-mueller	2021-04-09
a79ca2c	jens-daniel-mueller	2021-04-09
be095c6	jens-daniel-mueller	2021-04-09
eb827c9	jens-daniel-mueller	2021-04-07
857bad3	jens-daniel-mueller	2021-03-24
03b6009	jens-daniel-mueller	2021-03-23
555750f	jens-daniel-mueller	2021-03-23
a1d52ff	jens-daniel-mueller	2021-03-15
0bade3b	jens-daniel-mueller	2021-03-15
27c1f4b	jens-daniel-mueller	2021-03-14
af75ebf	jens-daniel-mueller	2021-03-14
5017709	jens-daniel-mueller	2021-03-11
585b07f	jens-daniel-mueller	2021-03-11

GLODAP_augmented %>%
  ggplot(aes(cstar_minus_mean)) +
  geom_vline(xintercept = 0) +
  geom_histogram(binwidth = 1) +
  scale_fill_viridis_c() +
  facet_grid(basin_AIP ~ data_source)

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
da61d1a	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
29444a1	jens-daniel-mueller	2021-08-05
42e80c0	jens-daniel-mueller	2021-08-04
48f6eed	jens-daniel-mueller	2021-08-04
81a46a4	jens-daniel-mueller	2021-08-03
b88c61b	jens-daniel-mueller	2021-08-03
a53656d	jens-daniel-mueller	2021-08-03
88f7356	jens-daniel-mueller	2021-08-02
d759279	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c9ccc00	jens-daniel-mueller	2021-07-22
910d64a	jens-daniel-mueller	2021-07-02
1cbf907	jens-daniel-mueller	2021-07-02
955f5dc	jens-daniel-mueller	2021-06-29
9480f94	jens-daniel-mueller	2021-06-29
d054c39	jens-daniel-mueller	2021-06-24
87e2739	jens-daniel-mueller	2021-06-24
4fcd066	jens-daniel-mueller	2021-06-24
7e1f407	jens-daniel-mueller	2021-06-10
2cbe18c	jens-daniel-mueller	2021-06-10
594ed9a	jens-daniel-mueller	2021-06-04
db7df0e	jens-daniel-mueller	2021-06-04
207339d	jens-daniel-mueller	2021-06-03
315710b	jens-daniel-mueller	2021-06-03
d37a85d	jens-daniel-mueller	2021-05-31
25bd183	jens-daniel-mueller	2021-05-26
62bd574	jens-daniel-mueller	2021-05-20
7c56c39	jens-daniel-mueller	2021-05-19
52e7583	jens-daniel-mueller	2021-05-12
969e631	jens-daniel-mueller	2021-05-12
d2a83bc	jens-daniel-mueller	2021-04-16
c0a47df	jens-daniel-mueller	2021-04-16
50290e8	jens-daniel-mueller	2021-04-16
b6fe355	jens-daniel-mueller	2021-04-16
81b7c6d	jens-daniel-mueller	2021-04-16
ddec5b7	jens-daniel-mueller	2021-04-15
099d566	jens-daniel-mueller	2021-04-14
bb44686	jens-daniel-mueller	2021-04-14
bf40480	jens-daniel-mueller	2021-04-13
9f31fe3	jens-daniel-mueller	2021-04-13
338dd3c	jens-daniel-mueller	2021-04-09
a79ca2c	jens-daniel-mueller	2021-04-09
eb827c9	jens-daniel-mueller	2021-04-07
857bad3	jens-daniel-mueller	2021-03-24
03b6009	jens-daniel-mueller	2021-03-23
555750f	jens-daniel-mueller	2021-03-23
83a13de	jens-daniel-mueller	2021-03-20
cf98c6d	jens-daniel-mueller	2021-03-16
a1d52ff	jens-daniel-mueller	2021-03-15
0bade3b	jens-daniel-mueller	2021-03-15
27c1f4b	jens-daniel-mueller	2021-03-14
af75ebf	jens-daniel-mueller	2021-03-14
5017709	jens-daniel-mueller	2021-03-11
585b07f	jens-daniel-mueller	2021-03-11

GLODAP_augmented %>%
  ggplot(aes(year, .resid)) +
  geom_hline(yintercept = 0) +
  geom_bin2d(binwidth = c(1, 1)) +
  scale_fill_viridis_c() +
  facet_grid(basin_AIP ~ data_source)

GLODAP_augmented %>%
  ggplot(aes(.resid)) +
  geom_vline(xintercept = 0) +
  geom_histogram(binwidth = 1) +
  facet_grid(basin_AIP ~ data_source)

8.1.2 Ranked RMSE

# calculate RMSE from augmented output per cruise
cruise_all <- GLODAP_augmented %>%
  group_by(cruise, data_source) %>%
  summarise(rmse = sqrt(c(crossprod(.resid)) / length(.resid))) %>%
  ungroup()

# rank RMSE
cruise_all <- cruise_all %>% 
  mutate(cruise = as.factor(cruise)) %>% 
  group_by(data_source) %>% 
  mutate(rank_rmse = rank(rmse)) %>% 
  ungroup()

cruise_out <- cruise_all %>% 
  filter(data_source == "obs",
         rmse > params_local$target_global_rmse_max)

GLODAP_out <- GLODAP_augmented %>% 
  filter(cruise %in% cruise_out$cruise)

ggplot() +
  geom_hline(yintercept = params_local$c_star_rmse_max) +
  geom_point(data = cruise_all,
             aes(rank_rmse, rmse)) +
  geom_point(data = cruise_out,
             aes(rank_rmse, rmse, col = cruise)) +
  facet_grid(. ~ data_source)

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
da61d1a	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
71546e4	jens-daniel-mueller	2021-08-06
29444a1	jens-daniel-mueller	2021-08-05
42e80c0	jens-daniel-mueller	2021-08-04
48f6eed	jens-daniel-mueller	2021-08-04
81a46a4	jens-daniel-mueller	2021-08-03
b88c61b	jens-daniel-mueller	2021-08-03
a53656d	jens-daniel-mueller	2021-08-03
88f7356	jens-daniel-mueller	2021-08-02
d759279	jens-daniel-mueller	2021-08-02
127b801	jens-daniel-mueller	2021-07-24
912d90e	jens-daniel-mueller	2021-07-23
2477316	jens-daniel-mueller	2021-07-23
c9ccc00	jens-daniel-mueller	2021-07-22
910d64a	jens-daniel-mueller	2021-07-02
1cbf907	jens-daniel-mueller	2021-07-02
955f5dc	jens-daniel-mueller	2021-06-29
9480f94	jens-daniel-mueller	2021-06-29
d054c39	jens-daniel-mueller	2021-06-24
87e2739	jens-daniel-mueller	2021-06-24
4fcd066	jens-daniel-mueller	2021-06-24
7e1f407	jens-daniel-mueller	2021-06-10
2cbe18c	jens-daniel-mueller	2021-06-10
594ed9a	jens-daniel-mueller	2021-06-04
db7df0e	jens-daniel-mueller	2021-06-04
207339d	jens-daniel-mueller	2021-06-03
315710b	jens-daniel-mueller	2021-06-03
d37a85d	jens-daniel-mueller	2021-05-31
25bd183	jens-daniel-mueller	2021-05-26
62bd574	jens-daniel-mueller	2021-05-20
7c56c39	jens-daniel-mueller	2021-05-19
52e7583	jens-daniel-mueller	2021-05-12
969e631	jens-daniel-mueller	2021-05-12
d2a83bc	jens-daniel-mueller	2021-04-16
c0a47df	jens-daniel-mueller	2021-04-16
50290e8	jens-daniel-mueller	2021-04-16
b6fe355	jens-daniel-mueller	2021-04-16
81b7c6d	jens-daniel-mueller	2021-04-16
ddec5b7	jens-daniel-mueller	2021-04-15
099d566	jens-daniel-mueller	2021-04-14
bb44686	jens-daniel-mueller	2021-04-14
bf40480	jens-daniel-mueller	2021-04-13
9f31fe3	jens-daniel-mueller	2021-04-13
338dd3c	jens-daniel-mueller	2021-04-09
a79ca2c	jens-daniel-mueller	2021-04-09
eb827c9	jens-daniel-mueller	2021-04-07
857bad3	jens-daniel-mueller	2021-03-24
03b6009	jens-daniel-mueller	2021-03-23
555750f	jens-daniel-mueller	2021-03-23
83a13de	jens-daniel-mueller	2021-03-20
cf98c6d	jens-daniel-mueller	2021-03-16
a1d52ff	jens-daniel-mueller	2021-03-15
0bade3b	jens-daniel-mueller	2021-03-15
27c1f4b	jens-daniel-mueller	2021-03-14
af75ebf	jens-daniel-mueller	2021-03-14
5017709	jens-daniel-mueller	2021-03-11
585b07f	jens-daniel-mueller	2021-03-11

8.1.3 Removed cruises

cruise_removal_ratio <- nrow(GLODAP_out)/nrow(GLODAP_augmented)*100

stats_MLR_cleaning_cruises <- tibble(
  cruise_removal_ratio = cruise_removal_ratio,
  cruise_removal_n = length(cruise_out)
)

stats_MLR_cleaning_cruises

# A tibble: 1 x 2
  cruise_removal_ratio cruise_removal_n
                 <dbl>            <int>
1               0.0445                4

cruise_removal_grid <- GLODAP_out %>%
  distinct(lat, lon, era, cruise)

map +
  geom_raster(data = GLODAP %>% distinct(lat, lon, era),
              aes(lon, lat), fill = "grey70") +
  geom_raster(data = cruise_removal_grid,
              aes(lon, lat), fill = "red") +
  facet_wrap( ~ era, ncol = 1) +
  labs(title = "Maps of removed cruises")

Version	Author	Date
cd8e0d5	jens-daniel-mueller	2021-08-06
15773a0	jens-daniel-mueller	2021-08-06
340d731	jens-daniel-mueller	2021-08-06
910d64a	jens-daniel-mueller	2021-07-02
1cbf907	jens-daniel-mueller	2021-07-02
9480f94	jens-daniel-mueller	2021-06-29
52e7583	jens-daniel-mueller	2021-05-12
969e631	jens-daniel-mueller	2021-05-12
eb827c9	jens-daniel-mueller	2021-04-07
857bad3	jens-daniel-mueller	2021-03-24
555750f	jens-daniel-mueller	2021-03-23
f155edd	jens-daniel-mueller	2021-03-23
83a13de	jens-daniel-mueller	2021-03-20
cf98c6d	jens-daniel-mueller	2021-03-16
a1d52ff	jens-daniel-mueller	2021-03-15
0bade3b	jens-daniel-mueller	2021-03-15
27c1f4b	jens-daniel-mueller	2021-03-14
af75ebf	jens-daniel-mueller	2021-03-14
5017709	jens-daniel-mueller	2021-03-11
585b07f	jens-daniel-mueller	2021-03-11

9 Write files

# select relevant columns
GLODAP <- GLODAP %>%
  filter(!(cruise %in% cruise_out$cruise)) %>%
  select(
    year,
    date,
    era,
    basin,
    basin_AIP,
    lat,
    lon,
    depth,
    data_source,
    gamma,
    gamma_slab,
    params_local$MLR_predictors,
    params_local$MLR_target
  )

GLODAP %>% write_csv(paste(
  path_version_data,
  "GLODAPv2.2020_MLR_fitting_ready.csv",
  sep = ""
))

co2_atm_tref %>%  write_csv(paste(path_version_data,
                                  "co2_atm_tref.csv",
                                  sep = ""))

tcant_3d %>%  write_csv(paste(path_version_data,
                             "projected_tcant_3d_tref.csv",
                             sep = ""))

tcant_zonal %>%  write_csv(paste(path_version_data,
                             "projected_tcant_zonal_tref.csv",
                             sep = ""))


stats_MLR_cleaning_cruises  %>%  write_csv(paste(
  path_version_data,
  "MLR_cleaning_cruise_removal_stats.csv",
  sep = ""
))

cruise_removal_grid  %>%  write_csv(paste(
  path_version_data,
  "MLR_cleaning_cruise_removal_grid.csv",
  sep = ""
))

sessionInfo()

R version 4.0.3 (2020-10-10)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: openSUSE Leap 15.2

Matrix products: default
BLAS:   /usr/local/R-4.0.3/lib64/R/lib/libRblas.so
LAPACK: /usr/local/R-4.0.3/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] seacarb_3.2.14  oce_1.2-0       gsw_1.0-5       testthat_2.3.2 
 [5] broom_0.7.5     lubridate_1.7.9 marelac_2.1.10  shape_1.4.5    
 [9] ggforce_0.3.3   metR_0.9.0      scico_1.2.0     patchwork_1.1.1
[13] collapse_1.5.0  forcats_0.5.0   stringr_1.4.0   dplyr_1.0.5    
[17] purrr_0.3.4     readr_1.4.0     tidyr_1.1.2     tibble_3.0.4   
[21] ggplot2_3.3.3   tidyverse_1.3.0 workflowr_1.6.2

loaded via a namespace (and not attached):
 [1] fs_1.5.0                 RColorBrewer_1.1-2       httr_1.4.2              
 [4] rprojroot_2.0.2          tools_4.0.3              backports_1.1.10        
 [7] utf8_1.1.4               R6_2.5.0                 DBI_1.1.0               
[10] colorspace_1.4-1         withr_2.3.0              tidyselect_1.1.0        
[13] compiler_4.0.3           git2r_0.27.1             cli_2.1.0               
[16] rvest_0.3.6              xml2_1.3.2               labeling_0.4.2          
[19] scales_1.1.1             checkmate_2.0.0          digest_0.6.27           
[22] rmarkdown_2.5            pkgconfig_2.0.3          htmltools_0.5.0         
[25] dbplyr_1.4.4             rlang_0.4.10             readxl_1.3.1            
[28] rstudioapi_0.11          farver_2.0.3             generics_0.0.2          
[31] jsonlite_1.7.1           magrittr_1.5             Matrix_1.2-18           
[34] Rcpp_1.0.5               munsell_0.5.0            fansi_0.4.1             
[37] lifecycle_1.0.0          stringi_1.5.3            whisker_0.4             
[40] yaml_2.2.1               MASS_7.3-53              grid_4.0.3              
[43] blob_1.2.1               parallel_4.0.3           promises_1.1.1          
[46] crayon_1.3.4             lattice_0.20-41          haven_2.3.1             
[49] hms_0.5.3                knitr_1.30               pillar_1.4.7            
[52] reprex_0.3.0             glue_1.4.2               evaluate_0.14           
[55] RcppArmadillo_0.10.1.2.0 data.table_1.13.2        modelr_0.1.8            
[58] vctrs_0.3.5              tweenr_1.0.2             httpuv_1.5.4            
[61] cellranger_1.1.0         gtable_0.3.0             polyclip_1.10-0         
[64] assertthat_0.2.1         xfun_0.18                RcppEigen_0.3.3.7.0     
[67] later_1.1.0.1            viridisLite_0.3.0        ellipsis_0.3.1          
[70] here_0.1

eMLR data preparation

Jens Daniel Müller

09 August, 2021

1 Version ID

2 Required data

2.1 GLODAP

2.2 Cant fields

2.3 CO2 atm and disequi

2.4 Surface ocean

2.5 Others

2.6 Convert to long format

3 PO₄* calculation

4 C*

4.1 Stoichiometric ratios

4.2 Calculation

4.3 Reference year adjustment

4.3.1 Join Cant fields

4.3.2 tcant at tref1

4.3.3 Revelle factor ratio

4.3.4 Alpha

4.3.5 tcant at tref2

4.3.6 Combine GLODAP + tcant

4.3.7 Merge GLODAP + atm. pCO₂

4.3.8 Calculation

4.4 Control plots

5 Isoneutral slabs

6 Rarefication

7 Observations coverage

7.1 Histograms

8 Cruise quality check

8.1 MLR models

8.1.1 Residuals

8.1.2 Ranked RMSE

8.1.3 Removed cruises

9 Write files

eMLR data preparation

Jens Daniel Müller

09 August, 2021

1 Version ID

2 Required data

2.1 GLODAP

2.2 Cant fields

2.3 CO2 atm and disequi

2.4 Surface ocean

2.5 Others

2.6 Convert to long format

3 PO4* calculation

4 C*

4.1 Stoichiometric ratios

4.2 Calculation

4.3 Reference year adjustment

4.3.1 Join Cant fields

4.3.2 tcant at tref1

4.3.3 Revelle factor ratio

4.3.4 Alpha

4.3.5 tcant at tref2

4.3.6 Combine GLODAP + tcant

4.3.7 Merge GLODAP + atm. pCO2

4.3.8 Calculation

4.4 Control plots

5 Isoneutral slabs

6 Rarefication

7 Observations coverage

7.1 Histograms

8 Cruise quality check

8.1 MLR models

8.1.1 Residuals

8.1.2 Ranked RMSE

8.1.3 Removed cruises

9 Write files

3 PO₄* calculation

4.3.7 Merge GLODAP + atm. pCO₂