Data cleaning, exploration and LSVI of HT31xx

[ToonHub]

I managed to run the script without any errors and the code is documented quite well . Good!

But I still have some comments and suggestions. See below.

[ToonHub]

Some suggestions for the table site_caractheristics:

• NA's for var: can be added quite easily based on the var_code
• if you use de 'Voorwaarde'-name in the LSVI-database for 'var' you do not have to change it afterwards when calculating LSVI. I would defintely change the name '31xx_lsvi_horizontalestructuur' and '31xx_lsvi_boom' as it is not clear wath they mean. Maybe also remove '31xx' in the var names?
• For some values that are not numeric it is not clear wath they mean (for example booev, bozi,...). You can add a column 'value_description' or just change the value names.
• I understand the use of is_above_LOQ for measuring depths. But I would not use is_below_LOQ in case of cover = '<1', because you also have cover values of 0, so you can quantify the cover below 1. I guess '<1' actually means that the cover is > 0 and < 1. You might use 0.5 for 'value_numeric' (as you do later in the lsvi-calculation) and use '<1' for 'value'

[ToonHub]

In the LSVI-calculation you use '1-out-all-out' for the aggregation. If you want to use the method that was used for the HR-reporting, you sould use Aggregatiemethode = "RapportageHR".

[ToonHub]

To make the code better readable I suggest to always start a new line after %>%

[anleyssen]

Suggested changes in HT31xx data preparation are made:

• add value_description for class variables,
• change 0-x-1 and <1 into 0.5 for value_numeric and change is_below_LOQ,
• replace NA in var and value_description

Some minor changes in HT31xx_LSVI (text edits) are also made. Can @ElsLommelen also review the code for LSVI-calculation?

[anleyssen]

The suggested changes of Els Lommelen are made in HT31xx LSVI. Other 'to do's' can't be solved at the moment; so @ToonHub, you can merge this branch with the master.