NEON Data Services Working Group (DSWG) Meeting 1
------------------------------------------------------------------------
November 22, 2010 12:00 pm Pacific
Participants
---------------
Jones, Peet, Hutchison, Schimel, Aulenbach, Ruhl, Vieglais, Gardiner, Palanisamy, Griffith, Burek, Parsons, Domenico, Habermann, Greenlee, Sangil, Guralnick, Erickson
Notes
--------
Background discussion
-----------------------------
Schimel -- overview and status report on NEON
-- Vision of NEON: provide env inf on critical questions, in most usable possible way
-- sites selected for representativeness, applicability to grand challenge questions
-- flow from sensors, people, through qa/calibration, through publication to user communities
-- substantial number of data sets come from outside the observatory
-- user model: science/education/decision maker using NEON information as seamlessly as possible
-- data from PIs, or LTER, or federal Inv and monitoring programs
-- about 120 high-level data products in initial incarnation (539 variables)
-- annual summaries
-- through assimilation models that produce gridded products across entire nation
-- observatory is about information, and making that information usable
-- DSWG is at heart: standards for products, how to capture provenance, assess usability of data
-- goal to make data as interoperable as possible for 539 variables spanning atmosphere to genomics
-- what are the key standards, best practices
-- what are the early, mid, and late project implementation priorities
-- now NEON is looking to move from proposal phase (8000) pages to working with community to make operational decisions
Questions:
Haberman: what is relation between 120 products and 539 variables
539 are the level 1 data variables
120 are the level 4 data products, assimilate the 539 variables
Mike: how many domains? RS to genomics? How many domains are we talking about
Schimel: less than a dozen, on the scale of 'atmospheric science'
Followup: how many schema are we dealiing with?
Schimel: grist for the committee, but probably 4-6?
Jones: NEON familiar enough
Haberman: familiar enough to get started
Griffith, Vieglais, Sangil
Scope of work discussion
-- review document on wiki
-- how extensive are the external data?
-- DEM, MODIS,
-- need provenance and full attribution for those
-- about 40 data sets
-- mainly in category of land use
-- also products they want to interoperate with -- not a strong handle on what these are
-- parsons: how related to LTER data?
-- schimel:
1. NEON to LTER as a provider to LTER sites -- should be easy to use in the LTER context as possible
2. NEON data activities will continue several products started at LTER, complementarity
-- most NEON data are 'observational', whereas a lot of LTER is on manipulative, process studies
-- needs interoperability
-- what are the high-priority target communities?
-- list isn't discrete
-- lots of communities at various levels of organization
-- disease and microbial ecology as partners, sometimes already part of LTER, others not
-- item for the group: what communities do we see as discrete and that should be addressed
-- could address it as different levels of interoperability
-- for some groups, metadata-only interoperability
-- for other key domains, deeper levels of integration
-- parsons: also identify other community portals that would be relevant, and distribution mechanisms [not so much the portals as the protocols and metadata standards that different portals use. In other words let those portals that want the data or metadata come get it]
-- user communities: who prioritizes them?
-- schimel
-- first community is scientific research community
-- others, such as decision makers are 2nd, 3rd, or 4th, etc.
-- educators at a variaety of levels (further out on the scale)
-- committees advice on segments of research community would be most useful
-- don't necessarily focus on education targets, etc, yet in the priority list
-- citizen science may be higher on the list
-- does GIS provide a crosscut across community?
-- possibly, but there are other paradigms (e.g., bioinformatics, girdded modeling, assimilation modeling, process study investigators focused on time series) also should be considered
-- pete Ruhl: who else should NEON interact with? e.g., ESIP,
Initial deliverables
-----------------------
-- Target user communities
-- focus on researchers
-- what are their current practices, activities -- we should produce a synopsis/overview of the landscape
-- pick a few to understand in greater detail
-- Metadata standards
-- Data standards
-- Provenance approaches
-- persistence of availability of data sets
-- citability of data sets (getting e.g., a DOI)
-- Data distribution protocols
-- services that NEON should support (e.g., OGC, OPeNDAP, etc)
-- Data Portals and aggregators
-- Recommended partnerships and interactions (e.g., ESIP, others) for standards, etc. (in each of above sections)
-- leverage existing activities (e.g., DataONE work on data citation)
Meta-considerations re: deliverables recommendation
-- need to highlight tradeoffs in above to help NEON with decisionmaking
-- Sangil: would like to see biographies for the committee
-- Wiki section for "Why am I here"
-- Sangil: has NEON started talking to target early data consumers-- what do they expect in terms of consuming data? A process to evaluate current metadata & data standards on the light of consumer expectations.
Action items
---------------
* Jones: doodle for future meeting times (standard slot, 3rd week of each month)
* Jones: load notes to wiki
* Jones: enumerate deliverable items on wiki
* Aulenbach: help work out connectivity issues for the wiki
-- contact Steve Aulenbach at NEON directly for your NEON DSWG wiki logon credentials
-- 720.746.4855
-- saulenbach@neoninc.org
* All: review the deliverables and make expansions/comments in preparation for next meeing
* All: review background data products for next meeting