|
This is version 2.
It is not the current version, and thus it cannot be edited.
[Back to current version]
[Restore this version]
- Want old datasets, old treatments/classifications of bats
- Has mappings (digital format) b/w names as in Wilson/Reeder with valid names 1993
- some geographic ranges - difficult with past names
- will find out what others do wrt identifying species from old data
- Nico: want to track species naming for a group back to Linneaus (she laughs)
- Interesting: Pipistrellus genus, red bat, Molossidae
- Our plan:
- Get data from Kate
- See what is missing
- How can we make it easy for her to enter missing links
- different type of migration
- insectivors
- ecological issues
- Minimal interaction - SMS is client of TOS - Shawn & Thau's ant demo showed problems
- They were modeling for ants - species is an instance of a class in OWL
- Info to be resolved can be done with database approach, difficult to scale up
- observations spread over thousands of species - in their approach load everything into memory
- If they contact server for each species, lots of overhead
- Matt thinks lots of overlap - Taxon is special case of SMS
- Susan - Overhead dealt with by resolving datasets ahead of time and stored/indexed
- Data integration - can't easily be done ahead of time
- We are further along with working system than SMS, so that may hamper integration
- Email SMS asking for OWL format for Taxon info
- JK: Where do we fit in?
- Niche modeling - search ecogrid for species occurrence data with name/species/concept overlaps
- Looking for name - may look for something precise or loosely defined
- Look for any number of taxa and return clustered groups of consistent taxa
- Want non-overlapping list of concepts for modeling?
- N-way mapping of concepts
- for mammals of NA, mammals w/in bounding box - list
- every item should represent single concept
- drag each item on canvas and user decides whether to use for analysis
- BEAM proposal that Nico sent out is good description of SEEK needs from Taxon
- Digir data - may need private aggreement to use collector/date though not expose
- EML - taxonomic metadata needs to be present, but usually in the data itself
- now, 3 distinct locations in EML, entire thing should be revisited
- relationship b/w EML and Biological Data Profile (BDP) FGDC data standards
- BEAM think about scenarios useful to Taxon
- Niche modeling case study (mostly museum data)
- Biodiversity case study (more ecological data) - is there documentation? Not yet
- Domain experts?
- Maybe Rob or Aimee go to KR/SMS/BEAM mtg - 7-10 March at UC Davis
- Engage SMS group before May
- ex: OWL output for Taxon info - get someone with more ecological background involved as well - Mark, Deana, Matt
- Next time we contact Shawn, ask Matt to join us
|
