[Paper Review] Open government geospatial data on buildings for planning sustainable and resilient cities

Filip Biljecki, Lawrence Zheng Xiong Chew+2 more|arXiv (Cornell University)|Jun 28, 2021

Geographic Information Systems Studies68 references32 citations

TL;DR

The paper inventories open government building data worldwide (140+ releases in 28 countries, over 100 million buildings) and analyzes accessibility, richness, quality, harmonisation, and actor relationships.

ABSTRACT

As buildings are central to the social and environmental sustainability of human settlements, high-quality geospatial data are necessary to support their management and planning. Authorities around the world are increasingly collecting and releasing such data openly, but these are mostly disconnected initiatives, making it challenging for users to fully leverage their potential for urban sustainability. We conduct a global study of 2D geospatial data on buildings that are released by governments for free access, ranging from individual cities to whole countries. We identify and benchmark more than 140 releases from 28 countries containing above 100 million buildings, based on five dimensions: accessibility, richness, data quality, harmonisation, and relationships with other actors. We find that much building data released by governments is valuable for spatial analyses, but there are large disparities among them and not all instances are of high quality, harmonised, and rich in descriptive information. Our study also compares authoritative data to OpenStreetMap, a crowdsourced counterpart, suggesting a mutually beneficial and complementary relationship.

Motivation & Objective

• Assess global availability of open government geospatial data on buildings across continents.
• Benchmark datasets on five dimensions: accessibility, richness, quality, harmonisation, and relationships with other actors.
• Identify gaps, best practices, and potential complementarities with OpenStreetMap and other data sources.

Proposed method

• Identify open government building datasets satisfying open data criteria and governmental authorship.
• Catalogue metadata per dataset: location, coverage level, creation/update year, format, and licensing.
• Assess each dataset against five dimensions using 13 qualitative/quantitative indicators.
• Cross-check datasets against OpenStreetMap imports to evaluate integration.
• Analyze attributes and geometry to determine richness and potential for 3D modelling.
• Synthesize findings into recommendations and best practices for data publication.

Experimental results

Research questions

• RQ1 What is the global coverage and accessibility of open government building datasets?
• RQ2 How rich and how complete are the available datasets in terms of attributes and geometry?
• RQ3 What is the data quality and how harmonised are the datasets across jurisdictions?
• RQ4 How do government data relate to and interact with OpenStreetMap and other actors?
• RQ5 What best practices and policy implications can improve open building data for urban sustainability?

Key findings

• Identified more than 140 open government building datasets spanning 28 countries, covering over 100 million buildings.
• Most datasets are freely accessible, but user experience issues and non-uniform download options hinder easy reuse.
• About half of datasets (53%) provide only 2D footprints; 47% include at least one attribute beyond geometry.
• Around 53% of datasets have attributes with building type being the most common; some datasets include height, address, or year of construction.
• Approximately 90%+ provide metadata on last update; half were created or updated within the last year; some datasets lack transparent update frequency.
• About 86% of datasets are in ESRI shapefile format; 40% are available in two or more formats; international standards harmonisation is limited (INSPIRE used in EU); geometric validity is high (less than 0.1% invalid geometries).
• Only about 25% of government datasets have been integrated into OpenStreetMap, indicating complementary opportunities and a potential for mutual benefits.
• There are notable data gaps in the Global South, and a strong need for cross-scale, harmonised data suitable for comparative studies.