How TERN Data Infrastructure works

How TERN Data Infrastructure works

Presentation by Tim ClancyTERN Director

Purpose

• Build and manage data infrastructure to provide public access to terrestrial ecosystem data.

• Facilitate open access to terrestrial ecosystem research data.

• Promote the culture of data sharing and re-use in ecosystem community.

Instruments + Sensors

Policy + Management

Analysis + Synthesis

Modelling

Data Searching

Data Sharing

Data Curation + Publishing

Data Storage

Processing + Analysis

Collection Methods

• TERN’s infrastructure for ecosystem science

Data Management ChallengesData heterogeneity: wide variety from different domain

• observation (human, in-situ sensor and satellite)Variety of scale: spatial

• point, plot, site, local government, state and continental scaleTemporal scale

• Varies from monthly, yearly and long-term observational spanning several decadesData formats

• CSV, NetCDF, Text description, Raster and Vector Metadata standards

• EML, ISO 19115 or 19139, custom metadata.Common data exchange format

• RIF-CS: feed to TDDP and ANDS RDAData archival

• Distributed across AustraliaAccessibility

• Adhere to TERN Data licensing Policy and framework• Enable access to citable data through DOI

Australian terrestrial ecosystem source data(Databases, repositories, data files, images, shapefiles)

Map layers(WMS)

Feature Data(WFS)

NetCDF(THREDDS)

Rich Contextual Info (ÆKOS) ISO 19115/9

(Geonetwork)

TERN DOI Minting service

OAI-PMH Harvester

TERN Data Catalogue Portal

Sear

ch A

PI

Exte

rnal

Pr

oces

ses

HTTP

Portal(s)HTML5, OpenLayers,

JavaScript, CSS

Service Interfaces, Metadata Interchange services, Transformation services, Business logic

Data Management Tool(s) (Morpho, SHaRED, ANZMET Lite)

Analysis and Synthesis (CoESRA)

Registry Interchange Format –Collections and Services(RIF-CS)

TERN Data Management: Key Elements

JSON

File system

Field Data Collections

EML(Metacat)

File system

DataOne Member Node

AuScribe App

TERN data licensing policy 2.0

• The least-restrictive licence/terms for all data made available through TERN.

• Data generated through TERN funding (“TERN data”) will be made freely and openly available by the relevant Facility, noting that:

Users will be required to attribute the source of the data; and Justifiable conditions protecting sensitivities of data will be

allowed.

• Updated in October 2015.

Schematic representation of TERN Data Licensing

Distributed infrastructure

Data Access: Across TERN

Access Data: TERN Data Discovery Portal

Portal.tern.org.au

Result Page

Fractional cover PV, NPV, Bare Soil

ANUCLIMATE 1.0

Logan river water quality

MODIS Grass curing Index

Biogeophysical Dataset collection

LTERN

Supersites

Ecology Data Collection

APPLICATION

Fire Management

Pre-processed MODIS fire burnt area satellite imagery

Vegetation Map andExpert elicitation

TERN’s impact on Terrestrial Ecosystem research data sharing

• Domain specific data management• Data and meta-entry tools• Metadata standards• Open standards for data delivery

• Flexible licensing policy• Links to national research data catalog• Ability to provide citable data (with DOIs)• Scalable and replicable infrastructure

TERN’s impact on ecosystem science and management

• Standardised data collection methodologies• New continental scale data products• Reduce duplication across jurisdictions• Improve knowledge for science to management• Promote collaboration and re-use of data

• higher return of investment for funding agencies

TERN Data Publication HighlightsOver 2000 data collections including:• Publish data from over 100,000 ecological sites;• Over 40 continental scale remote sensing data products;• Over 30 continental scale soil and landscape attributes data;• Coastal ecosystem datasets including national seagrass,

beach observation and water quality;• Continental scale data on climate variables at 1 km spatial

resolution;• Half-hourly time-series flux data from towers across

Australia.

Moving forwards – sustaining long term science

•Global shift to collaborative data , algorithms and participatory resources:

Methods 1 – BCC Virtual Lab

Methods 2 – CoESRA Pilot (Guru et al 2015)

Logo

Conclusion

• Significant collection, collation of ecosystem data

• Large institution databases are available on open access.• State government vegetation survey data

• More details during the course of the day.

International Partners

TERN is supported by the Australian Government through the National Collaborative Research Infrastructure Strategy and the Super Science Initiative

More information

TERN website: www.tern.org.au

Thank yout.clancy@uq.edu.au