TDWG2026
Home / Talks / SAL C
talk · Friday 25 September · 10:40–⁠10:55 · SAL C

Building a Living Publication Workflow for Long-Term Social-Ecological Monitoring Data: The Taiwan LTSER Experience

Jhu-Jyun Jhang · From Standards to Implementation: Connecting Observation Data in Asia to GBIF Infrastructure

Recording time 2:55:00–3:02:43Open on Vimeo ↗

Summaries and key moments are machine-produced. Check the recording for the speaker’s exact words.

The short versionEmbedding Darwin Core alignment as automated checkpoints in a repository-based pipeline keeps continuously updated monitoring data flowing to GBIF.

Overview

What this was about

Jhu-Jyun Jhang presented a semi-automated 'living' publication workflow for Taiwan's Long-Term Social-Ecological Research (LTSER) network, which collects ecological, environmental and social data. Research teams fill a co-designed data template and upload to the depositar repository; a robot polls the depositar API daily, validates updated datasets against the template, emails contacts on failure, loads valid data into the LTSER portal, maps them to Darwin Core, syncs to the TaiBIF IPT and publishes to GBIF and TBIA. Four datasets have been published this way so far.

Why it matters. Long-term monitoring data are rarely republished after updates; an automated pipeline lowers the burden on researchers and keeps GBIF copies current.

Key ideas

In the room

  • Taiwan LTSER is a site-based monitoring project combining ecological, environmental and social data to understand climate change and human impacts.
  • Continuously growing datasets make one-time publication inadequate; repeated extraction and Darwin Core alignment is costly, so data risk never reaching biodiversity infrastructures.
  • The workflow has three phases: research teams preparing data, an automated robot/script phase, and publication to GBIF and TBIA.
  • An LTSER data template co-designed with each team specifies field name, definition, data type, controlled values, examples and requirement status, easing later Darwin Core mapping.
  • depositar provides public data/metadata APIs, multiple formats and persistent ARK identifiers.
  • A robot checks depositar daily at 3 a.m.; updated datasets are validated, invalid ones trigger an email to the team contact, valid ones go to the LTSER portal, are mapped to Darwin Core and synced to the TaiBIF IPT.
  • Four datasets have been published through the workflow.
Jump in

Notable moments

In their words

Transcript

Automatically generated captions can contain mistakes, especially in names and technical terms. Times are relative to the room recording.

Read the transcript ↓
Loading transcript…
Keep exploring

Related talks.

See the map ↗
talk · 25 Sept · 11:25–⁠11:40 · SAL C

From sampling-event datasets to the Humboldt Extension: a TaiBIF assessment and pilots for long-term monitoring in Taiwan

Giving monitoring-data providers immediate site-level trend feedback, based on a minimal set of Humboldt terms, can motivate adoption better than asking for the full extension.

Jerome Chie-Jen Ko
talk · 25 Sept · 11:10–⁠11:25 · SAL C

Reporting structure and sampling-bias correction in citizen-science bird data across Asian cities

Abundant citizen-science records do not guarantee inference support; reporting structure can distort richness hotspots and urban-biodiversity relationships unless it is diagnosed and controlled.

Yeshan Qiu
talk · 25 Sept · 11:18–⁠11:30 · SAL A

Reliable LLM-assisted curation of ecological survey data: a deployed agent bridging field collection and data curation

An LLM agent that proposes findings for curator approval, with confirmed patterns turned into deterministic rules, catches plausibility errors that schema validation misses.

Andrew Tokmakoff
Programme abstract

Building a Living Publication Workflow for Long-Term Social-Ecological Monitoring Data: The Taiwan LTSER Experience

Jhu-Jyun Jhang

Long-term social-ecological monitoring data are growing, continuously updated resources that require living publication workflows rather than one-time releases. In long-term social-ecological research networks, biodiversity observations are often collected with environmental, social, and site-based monitoring data. This makes it challenging to extract biodiversity relevant records and routinely transform, validate, and publish them through GBIF compatible infrastructures.

Taiwan LTSER is developing a semi-automated workflow to help continuously updated monitoring datasets move toward Darwin Core alignment. The workflow uses depositar, a research data repository for managing and sharing datasets with metadata, to bridge local data management practices and downstream biodiversity data publication infrastructures.

From this repository-based entry point, scheduled checks detect dataset updates and evaluate whether records can be aligned with Darwin Core. In this approach, biodiversity data standards are implemented as workflow checkpoints rather than tasks left to individual researchers. Datasets that require revision can be returned to project contacts for correction, while datasets that meet the requirements are first presented on the LTSER portal and then prepared for further publication through TaiBIF IPT and GBIF.

The Taiwan LTSER experience shows how a living publication workflow can reduce manual work, lower technical barriers, and connect distributed long-term monitoring data to global biodiversity infrastructures. It demonstrates that the key implementation challenge is not only adopting data standards, but embedding them into practical, repeatable workflows that fit real research settings.

Submitted to the conference programme before the talk, so the talk itself may differ.