add read abiotic ranges function

[hansvancalster]

A function to read the data from https://zenodo.org/records/10533792.

This is a quick and dirty implementation - so a good review round will be needed.

@florisvdh I haven't updated NEWS or bumped version because I imagine that you may prefer setting up a development branch first? Otherwise I can add it to this branch, just let me know.

[florisvdh]

@hansvancalster thanks for giving this a start! As already talked about, we'll look into this further together. I plan to come back to it later this week.

The target branch should be dev_nextrelease (not main), and it has got updated recently. So you need to update the target branch of this PR (I suppose you can, otherwise I can) and pull dev_nextrelease into read_far.

If you want a clean start instead, then, after pulling dev_nextrelease, you can either start a new branch from there, cherry-pick ebafc10 and start a new PR, or update the existing branch by rebasing read_far onto dev_nextrelease, then force-push here.

[florisvdh]

> test <- read_abiotic_ranges()
> glimpse(test)
Rows: 1,149
Columns: 18
$ Habitatsubtype        <chr> "1310_pol", "1310_pol", "1310_pol", "1310_zk", "1310_zk", "1310_zk", "1310_zk", "1310…
$ Subtype               <chr> "", "", "", "", "", "", "", "Suaedetum", "", "", "Sagino maritimae-Cochlearietum", "C…
$ Milieucompartiment    <chr> "Bodem", "Grondwater", "Lucht", "Grondwater", "Grondwater", "Inundatiewater", "Inunda…
$ Variabele             <chr> "Chloride", "Chloride", "Stikstofdepositie", "Chloride", "Gemiddelde voorjaarsgrondwa…
$ Afkorting             <chr> "Cl", "Cl", "N-depositie", "Cl", "GVG", "Hoogte t.o.v. getij", "Overstromingsfrequent…
$ Eenheid               <chr> "mg/l", "mg/l", "kg N/ha/j", "mg/l", "m - mv", "m", "%", "%", "kg N/ha/j", "m - mv", …
$ Toetswijze...bepaling <chr> "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "…
$ Teken                 <chr> "OG - BG", ">", "<", ">", "OG - BG", "", "OG - BG", "OG - BG", "<", "OG - BG", "OG - …
$ Abiotisch.bereik      <chr> "0 - 70", "3000", "23", "3000", "-0,13 - 0,03", "GHW", "50 - 85", "41 - 61", "23", "0…
$ Waarde.1              <chr> "0", "3000", "", "3000", "-0,13", "GHW", "50", "41", "", "0,22", "3000", "300", "", "…
$ Waarde.2              <chr> "70", "", "23", "", "0,03", "", "85", "61", "23", "0,41", "10000", "3000", "21", "", …
$ WaardeNum1            <chr> "0", "3000", "", "3000", "-0,13", "", "50", "41", "", "0,22", "3000", "300", "", "300…
$ WaardeNum2            <chr> "70", "", "23", "", "0,03", "", "85", "61", "23", "0,41", "10000", "3000", "21", "", …
$ N.gunstig             <int> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, N…
$ Status                <chr> "Lk", "Lk", "Lg", "Lk", "Lk", "Lk", "Lk,Lk", "Lk", "Lg", "Lk, Lk", "Lk", "Lk", "Lg", …
$ Referentie            <chr> "Schamin\xe9e et al. 1998", "Ertsen et al. 1998", "ANB 2015", "Ertsen et al. 1998", "…
$ Opmerking             <chr> "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "…
$ Wijziging             <lgl> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, N…

Dropping here some thoughts to discuss live.

• use applicable elements of other n2khab reading functions like read_habitatsprings() as a template (but don't use .data in tidyselect expressions, it's deprecated). This explains several of below.
• explicitly distinguish a far data source (for read_far()) (or favenv)
  • using directory 10_raw/far
  • explicitly referring the version in the function
  • adding it into our draft overview
  • change version name in Zenodo to e.g. far_v1 on Zenodo
  • preferably also include this data source in the n2khab-data-raw collection community on Zenodo, referred by the n2khab package
  • preferably provide both English & Dutch metadata + English title in the Zenodo page (cf https://zenodo.org/records/6652962)
• should we use 'abiotic' or 'environmental'? TBD
• downloading only when missing (actually that is a framework which we plan to implement in a general way, using a built-in dataset of all data source versions and their DOIs)
• style & structure of the result, including column names: https://inbo.github.io/n2khab/CONTRIBUTING.html#coding-tools
• consider the option to return either English or Dutch content, e.g. as in read_types(lang = ), which makes use of the namelist data source.
• use factors where appropriate (many of these variables), preferably using a sensible level order. Some of these should be interconnected, e.g. the environmental variable & its abbreviation
• match column names and factor levels with other objects; the first three variables should be aligned with output from read_types(), read_schemes()
• empty strings or NAs?
• units:
  • don't use Dutch unit abbreviations in the result (and I would recommend: the data source), e.g. 'j' is now used for year; also 'm-mv' is Dutch.
  • actually 'm-mv' is not a real unit. It is 'm', but using a local vertical CRS where zero is the soil surface level and with a declared direction. This is part of the environmental variable's definition.
• use UTF-8 for strings where this isn't the case (e.g. Referenties above). See e.g.

  n2khab/R/read_soilmap.R

  Lines 383 to 384 in 8be10bb

  	soilmap %>%
  	convertdf_enc(from = "latin1", to = "UTF-8") %>%

• optionally consider something as in read_habitatquarries() regarding the literature references