Revised the script ‘doi:10.18167/DVN1/2EHEQT.R’

[smkuhlani]

Revised the script ‘doi:10.18167/DVN1/2EHEQT.R’, in pending to the main folder /scripts/conservation_agriculture. 1-revised PK ammounts, based on literature search. Extrapolated fertilizer PK proportions. Assessment shows N, as 11%. Most common rice fertilizer with 11% N in Madagascar is 11:22:16 (NPK). https://agritrop.cirad.fr/539499/1/document_539499.pdf

[rhijmans]

Thanks.

• Please do not move scripts around like this because then we cannot see what changes were made. You can just leave them in pending

• The "treatment variables" are described as "site x management x soil x season"; but we do not add location and season (time and space; because they are always treatments in aggregated data). So that leaves management and soil, which became landscape_position;land_prep_method;OM_amount. OM_used is redundant and (less informative than) with OM_amount. It is, however, not clear to what extent these actually are (experimental) "treatments". There are 700 to 1150 fields per year. This looks more like a survey (perhaps with a control group). There is no metadata, unfortunately.