Update Belgian occurrence indicators based on corrected cubes

[damianooldoni]

Run occurrence based indicators on corrected cubes:

• Occurrence cubes for non-native taxa in Belgium and Europe: https://zenodo.org/records/10527772
• Occurrence cubes at species level for European countries: https://zenodo.org/records/10528506

This means running all Rmds in ./src from 5 to 12.

[soriadelva]

Update: the links to the corrected cubes on zenodo have been added where necessary. However, two errors occurred

• In script 09_define_overlay_grid_belgium_with_protected_areas.Rmd: The download of the Natura2000_end2019_Shapefile fails (a folder is created on the local computer but remains empty). It goes wrong somewhere in the following lines:

  indicators/src/09_define_overlay_grid_belgium_with_protected_areas.Rmd

  Lines 61 to 71 in 9e9668d

  	if (!file.exists(path_natura2000)) {
  	options(timeout = 60*30) # 30min max allowed for downloading
  	temp <- tempfile()
  	download.file(
  	"https://cmshare.eea.europa.eu/s/n5L8Lrs9aYD775S/download",
  	temp,
  	cacheOK = FALSE
  	)
  	unzip(temp, exdir = path_natura2000)
  	unlink(temp)
  	}

• In script 07_occurrence_indicators_modelling.Rmd the following error occurs when running the trias::apply_gam function:
  Error in map2(): ℹ In index: 6. Caused by error in dplyr::select(): ! Can't select columns that don't exist. ✖ Column smooth doesn't exist.
  This same error has also appeared in another repo that uses the apply_gam function: see Tests of package (via devtools::check() ) indicates an error and warnings trias#113

[damianooldoni]

First problem solved via c6e5763

[soriadelva]

@damianooldoni I tried to rerun the 07_occurrence_indicators_modelling.Rmd script but still encountered the same error linked to the apply_gam function (column smooth doesn't exist). It has likely something to do with how I loaded the updated version of the trias package on my local computer (I fetched the main branch of trias-project/trias and did "update from main" in Github Desktop for the current branch linked to issue #128 of trias-project/indicators). As that still gave an error, I updated the trias package on my local computer but that also didn't work. Do you know if I still need to do something else? Thanks in advance!

[damianooldoni]

Hi @soriadelva. The trias branch with the update was still not merged to main. That was indeed my task to do after your review. I have just merged to main. You could now via GitHub Desktop:

1. go to trias repository
2. check you are on main branch
3. click on "fetch origin"
4. Pull the changes
5. Open tias RStudio project related to trias R package
6. Run devtools::install()
7. Close it
8. Open indicator R project or close it and open it again.
9. Run library(trias)
10. Check you have the right package version by running devtools::session_info(). If you see a line with trias * 2.1.0 it's ok.

So, in general, it's just a question of checking versions of a package. This is a nice example why it's important to apply versioning in package development 😄

Please, don't hesitate to contact me here or via chat if something is not working or not clear.

[soriadelva]

Hi @soriadelva. The trias branch with the update was still not merged to main. That was indeed my task to do after your review. I have just merged to main. You could now via GitHub Desktop:

1. go to trias repository
2. check you are on main branch
3. click on "fetch origin"
4. Pull the changes
5. Open tias RStudio project related to trias R package
6. Run devtools::install()
7. Close it
8. Open indicator R project or close it and open it again.
9. Run library(trias)
10. Check you have the right package version by running devtools::session_info(). If you see a line with trias * 2.1.0 it's ok.

So, in general, it's just a question of checking versions of a package. This is a nice example why it's important to apply versioning in package development 😄

Please, don't hesitate to contact me here or via chat if something is not working or not clear.

Thanks @damianooldoni, this worked! I ran all scripts and knit the respective .Rmd files. Before I generate a Pull Request, do you want me to commit all the files that have been overwritten by the changes?

[damianooldoni]

Did you remember to delete all existent images before generating the new ones?

Yes, you can push the changes in some commits and then generate a Pull Request 👍 Thanks a lot.

[soriadelva]

Did you remember to delete all existent images before generating the new ones?

Yes, you can push the changes in some commits and then generate a Pull Request 👍 Thanks a lot.

oh no, I didn't! 😓 I'll run it again!!