Metadata for New Zealand's National Vegetation Plot Databank

1
Metadata for New Zealand's National Vegetation
Plot Databank

• Nick Spencer and Susan Wiser
• Landcare Research
• New Zealand

2
What is NVS?

• NVS (National Vegetation Survey) New Zealands
  largest archive facility for plot-based
  vegetation data

http//nvs.landcareResearch.co.nz
3
NVS - coverage

• Best in grassland and indigenous forest
• Collection intensity has varied over 50 years
• 14 000 permanent and
• 52 000 relevé plots
• NVS has many uses

4
Why metadata management?

• In the past good for organising data
• Expanding content and function makes metadata
  critical
• e.g. Kyoto protocol reporting
• Metadata system redeveloped to meet new demands

5
What is metadata?

• Metadata is information about information
• Who, What, Where, When, Why and How

6
Consequence of missing metadata

• Knowledge about a dataset is lost overtime

Time of publication
From Michener et al (1997)
Specific details are lost rapidly e.g. Dates
General details are lost through time
Retirement or career change makes access difficult
Accident may destroy data or documentation
Death of investigator and loss of remaining
records
Time
7
Why is metadata useful?

• Search and locate datasets
• Assess suitability of use
• Reduces the effort required to use data
• metadata leads to better information efficiency
• (Michener et al 1997)
• Caveat...
• A balance needed
• more metadata means less research
• (Michener et al 1997)

8
Recent developments

• Goals
• Comprehensive
• Standards based
• Versatile
• Approach
• 1. XML based storage structures (Schema)

9
What is XML?

• eXtensible Mark-up Language
• Similar to HTML but consists of user-defined
  tags to structure textual information
• Promotes universal data access
• Machine and human-readable
• Open standard
• Written in plain-text (ASCII)

10
Recent developments

• Goals
• Comprehensive
• Standards based
• Versatile
• Approach
• 1. XML based storage structures (Schema)

11
Recent developments

• Goals
• Comprehensive
• Standards based
• Versatile
• Approach
• 1. XML based storage structures (Schema)
• 2. Separate the metadata and data systems
• (see the demonstration following this talk)

12
Developing the schema

• Looked to external metadata standards
• and profiles
• ISO 19115 Geographic metadata standards
• DC Dublin Core
• EML Ecological Metadata Language
• Adopted universal elements

13
Our Metadata Schema

•  
• 34 primary metadata elements 105 distinct
  sub-elements
• 68 match with source standards
• Grouped broadly as
• Identity (title, Id)
• Content (information types, methods)
• Context (location, time, purpose)
• Admin (ownership, access, availability, status)

14
Notable features of our schema

• Resources or related material
• Internal (e.g. a child)
• Managed (e.g. photographs)
• External (e.g. bird counts)
• Metadata containers and versioning  

15
Outcomes

• Improved accessibility and consistency
• XML document approach
• Portable, flexible and extendable
• Readily reformated for different uses
• (e.g. web, text, apps)
•         
• But
• Few mandatory metadata elements
• Relational database ? structured XML
• XML tools and languages are less familiar
• c.f. SQL (20 year standard)

16
Acknowledgements
Data entry and administration Michelle Breach
Plant ecologists Peter Bellingham Susan
Wiser Larry Burrows Rob Allen
Dept. of Conservation Liaison Elaine Wright
IT strategists and developers Jerry Cooper
Nick Spencer Mark Fuglestad
Funded by Foundation for Research Science
Technology Department of Conservation Terrestrial
Freshwater Biodiversity Information System