TDWG2026
Home / Talks / Aula
talk · Monday 21 September · 15:00–⁠15:15 · Aula

Using automation to consolidate biodiversity occurrence data in South African National Parks

Judith Botha · Monday plenary programme

Recording time 1:09:28–1:24:45Open on Vimeo ↗

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

The short versionAutomating the harvest, integration and summary of biodiversity data lets park staff get evidence-based species lists and dashboards without laborious manual compilation.

Overview

What this was about

Judith Botha presented the SANParks Biodiversity Information Management System (BIMS), built with the Freshwater Research Centre and technical partner Kartoza. It brings together GBIF occurrence data, SANParks monitoring data and data captured from literature for South Africa's 21 national parks. The system produces annotated species checklists (needed for five-yearly park management plans), summary dashboards, Red List status and Red List Index trends, and climate station dashboards. Harvesting, taxonomic checks against the GBIF backbone, value assignment and dashboard generation are all automated.

Why it matters. This shows aggregated biodiversity data and standards reaching conservation managers and rangers as practical decision-support tools, and replacing the manual work behind statutory species lists.

Key ideas

In the room

  • SANParks manages 21 national parks (about 4.4 million ha terrestrial plus about 370,000 ha of marine protected areas). Kruger National Park is about 2 million ha and turned 100 this year.
  • South Africa has about 67,000 animal species, many poorly known and highly endemic.
  • Park management plans every five years need updated species lists, which used to be compiled by hand from literature.
  • BIMS harvests species and occurrence data from GBIF (occurrences for each reserve plus a buffer and catchment) and adds unpublished literature, databases, books, reports and theses captured by dedicated staff.
  • The GBIF taxonomic backbone keeps names to accepted names.
  • Users filter by taxon, space, time, Red List status, origin and endemism, and download annotated checklists (Excel, PDF, CSV). The checklists include coordinate uncertainty and precision, record counts, most recent record date, and an include/exclude flag for known misidentifications.
  • The dashboards show endemism, origin and conservation status. A Red List view uses SANBI's Red List Index by taxonomic group, and a climate dashboard summarises station data by time period.
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 · 24 Sept · 14:50–⁠15:05 · SAL B

Biodiversity data at your fingertips: open-access freshwater biodiversity information systems for South Africa and Africa

Co-designed, open and interoperable freshwater information systems like FBIS turn fragmented records into decision-ready information, and multi-tenancy helps keep them sustainable.

Helen Dallas
talk · 24 Sept · 09:47–⁠09:59 · FORUM

Architecting the National Biodiversity Information System: Standards-Driven Workflows for South Africa’s Botanical and Zoological Data

SANBI sustains NBIS through a statutory mandate plus a roadmap of diversified funding, partnerships and demonstrated return on investment.

Fhatani Ranwashe
talk · 24 Sept · 09:12–⁠09:25 · SAL C

Digitize, Standardize, Mobilize: Wildlife Biodiversity Biobanking with Specify

Specify's hierarchical trees and attachments turned a fragmented wildlife biobank into a traceable extended-specimen data system.

Kim Labuschagne
Programme abstract

Using automation to consolidate biodiversity occurrence data from multiple sources to assist conservation decision making in South African National Parks.

Judith Botha

South African National Parks manages 21 National Parks within South Africa that includes more than 4 million hectares spanning both the Marine and Terrestrial environments. To support evidence-based decision making at park level data from multiple sources need to be consolidated allowing users and park managers to easily access this consolidated data through dashboards. Automation and AI allows consolidation of open access data from sources outside of the organisation, as well as open access data from parks like collection data and sensitive park specific data to be viewed and used together by decision makers. APIs allow automated harvests of data from platforms like GBIF as well as information around conservation statuses of species from IUCN sites which is required for managing species at park level. Additional features allow integration of the data using selection features like selecting a certain park/area, taxonomic group, time frame or conservation status without any prior knowledge of GIS or data integration needed. Further features allow a user to download climate data for a particular area, create species lists and obtain additional information and graphs on the conservation status of taxonomic groups in these areas and how these have changed over time. This presentation will showcase this system and highlight how it uses AI and automation to generate easily accessible evidence.

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