2013-01-18 conference call¶

Martha's summary¶

• Jim - look into difficulty/time required to add county etc. coordinates and have GNRS: See /home/bien/inputs/.geoscrub/county_centroids/county-centroids.csv on vegbiendev
• Aaron - point Bob to the more intuitive view of the data to validate: validation views have been improved
• Aaron - mention to Peter to include MO cultivated field and specimen description data in the refresh
• Brad - connect Aaron and Jim Clark at Duke's future postdoc who has FIA expertise. Keep Brad/Bob in the loop on discussions. Determine which FIA fields are needed for BIEN. Kai contacted, but not available yet
• Brad - send/post updated development timeline: see timeline.2013.xls .

Timeline¶

*timeline.2013.xls*

• Updated with Brad's changes

To do¶

Analytical DB¶

1. separate specimens view

Validation¶

1. create Madidi re-export for Peter Jorgensen
2. create CVS, UNCC exports for Bob Peet
3. validate FIA
   • validation contact: Jim Clark's postdoc (@ Duke University), who Brian met at the International Biogeography Conference
     1. Brad will contact Brian to get name of contact: this is Kai Zhu .
     2. Bob will look for other postdocs working with FIA
   • what fields we need from FIA to be able to use it
   • how to deal with swapped geocoordinates, etc.

Data refresh¶

1. mention to Peter Jorgensen that MO refresh needs cultivated field and specimen description
2. refresh FIA

Range models¶

1. Re-run all range models in March

Attribution¶

1. attribution by April
2. collect contributor data use policies
3. putting together conditions of use for a data download

Download tracking¶

1. user DB for BIEN 3
   • potentially very time-consuming
2. automated workflow to create extracts for data providers

Georeferencing¶

1. each geocoordinate needs error bar (uncertainty)
2. coordinates table needs coordinateSource field to indicate where coordinate came from
   • closed list for provided by data provider or added by BIEN
3. add coordinates based on political division (esp. county): The centroids have been added to the analytical DB, which will be used when there is a county but no coordinates. However, note that the join does not yet use scrubbed county names, because geoscrub_output only has rows for data with both placenames and coordinates.
   1. county -> centroid/radius lookup table: See /home/bien/inputs/.geoscrub/county_centroids/county-centroids.csv on vegbiendev
      • for uncertainty, use longest dimension of bounding box, divided by 2?
      • Jim will look into this next week: See /home/bien/inputs/.geoscrub/county_centroids/county-centroids.csv on vegbiendev

Exchange schema¶

1. need template to give to data providers
   • hosted on BIEN or TDWG website
2. push exchange schema at next TDWG meeting

Long-term¶

1. replicate DB at iPlant