2026-09-02 –, Conference Management Room5
Official statistics lack impact without spatial context. This talk presents UNICEF’s open-source stack - GeoRepo, GeoSight, and SDMXConnector - enabling seamless integration of SDMX data with harmonized boundaries. We address key challenges and show how to transform complex indicators into map-ready insights for better decision-making and GeoAI readiness.
Official statistics are critical for decision-making - but without a spatial dimension, their full value often remains untapped. Maps transform tabular indicators into actionable insights, revealing geographic inequalities, trends, and hotspots that are otherwise difficult to detect. Yet, integrating statistical data - especially from SDMX (Statistical Data and Metadata eXchange format) into geospatial workflows remains complex and inaccessible for many users.
This talk presents UNICEF’s open-source approach to bridging that gap through a modular geospatial stack: GeoRepo, GeoSight, and a newly developed SDMX Connector. Together, these tools enable seamless integration of official statistics with harmonized administrative boundaries, making it easier to explore, visualize, and analyze subnational, multi-indicator time series data.
We begin by outlining the key challenges of geo-enabling SDMX data: matching statistical indicators with the correct administrative boundaries, handling evolving geographies over time, transforming multidimensional datasets into GIS-ready formats, and managing time-series complexity. We argue that sharing reference area codes alone is insufficient - true interoperability requires access to consistent, versioned geometries.
Next, we introduce GeoRepo, UNICEF’s open-source repository of global administrative boundaries, offering versioning, smart matching, and scalable APIs, and GeoSight, an open-source platform designed for visualizing statistical indicators alongside contextual geospatial layers. Building on this foundation, we showcase the SDMX Connector - developed in collaboration with Harvard Tech for Social Good - which allows users to directly browse, filter, preview, and import data from an SDMX registry into a geospatial environment.
By lowering technical barriers, this ecosystem empowers analysts, policymakers, and developers to move from static tables to dynamic, map-based insights. It also lays the groundwork for GeoAI applications by making statistical data spatially explicit and machine-readable.
https://github.com/unicef-drp/GeoSight-OS
https://unicef-drp.github.io/GeoSight-OS-Documentation/user/manual/home/
This talk highlights several open-source projects developed and maintained by UNICEF and partners:
1. GeoSight - an open-source geospatial platform for visualizing subnational, multi-indicator time-series data
2. GeoRepo - an open-source repository of harmonized global administrative boundaries with versioning and APIs
3. GeoSight SDMX Connector - an open-source extension enabling direct integration of SDMX data into geospatial workflows
Together, these projects form a modular, open ecosystem for bridging official statistics and geospatial intelligence.
I make my conference contribution available under the CC BY 4.0 license. The conference contribution comprises the abstract, the text contribution for the conference proceedings, the presentation materials as well as the video recording and live transmission of the presentation:
Jan Burdziej is a Geospatial Lead in UNICEF’s Computational and Geospatial Intelligence Team, where he leads the development of a corporate geospatial infrastructure using open-source and commercial technologies. With over 20 years of experience, he has delivered GIS solutions across UN agencies, national statistical offices, environmental organizations, and the private oil & gas sector. He holds an MSc in GIS from Salzburg University and a PhD in Geography from NCU. Outside work, he enjoys mountain hiking, running, and building wooden furniture for his children.