search

ANZSoilData / def-au-asls-substrate

Machine-readable representation of the classifiers described in chapter 9 Substrate, by R.C. McDonald and R.F. Isbell, in Australian soil and land survey field handbook (3rd edn)
0 stars 0 forks source link

challenge mapping substrate results #1

Open meganrwong opened 2 years ago

meganrwong commented 2 years ago

Here are some notes from an experience mapping one dataset. Perhaps not directly relevant to encoding YB as is, but may be of interest

We found mapping results relating to substrate difficult

What we contemplated - could we use CGI terms where YB where not available if we were confident we could apply the CGI terms correctly and use YB terms as they exist. Someday, a group of experts can help map how YB relate to CGI terms?

But, we decided to map all this as free text under obsv property of ‘substrate material’. This was because data reported as ‘parent material’ were a mix of lithological types of rock or unconsolidated material, or substrate mass genetic types (eg alluvium).

dr-shorthair commented 2 years ago

That would really be a business decision. The vocabs are typically provided 'as is' for use if they are suitable. If there are gaps then you have a couple of options:

  1. Choose a better vocabulary
  2. Get the preferred vocabulary fixed
  3. some kind of hybrid approach (which is what you are proposing, by mixing values from more than one source)
  4. ...