Disagreements in estimates of urban land from global maps: Implications for assessments of urban climate risks

The advancement of high-resolution satellite remote sensing and computational techniques has significantly improved our ability to detect global urban land cover and its evolution over time and space. These technological breakthroughs have led to the creation of various global urban land cover datasets, which are essential for assessing climate risks and understanding urbanization patterns in both observational and modeling frameworks in an increasingly urban world. However, despite their importance, these datasets often exhibit substantial discrepancies due to differences in urban definitions, classification methodologies, and data sources. These inconsistencies can lead to varying estimates of the extent and rate of urban expansion, and how they covary with various evolving hazards, posing challenges for researchers and policymakers who rely on these datasets for climate and urban planning studies.

Through an analysis of widely used, current-generation datasets, we observe a significant increase in global urban land area, which nearly tripled between 1985 and 2015. However, there are large discrepancies in the estimates of urban land across datasets from local to regional to continental scales. Interestingly, the largest divergences are seen for the most recent years on inclusion of the newly released 10 m resolution products, partly due to their ability to better resolve urban facets. This rapid urban expansion has profound implications for climate systems, including localized urban warming, increasing urban flood risks, and changes in regional atmospheric patterns. We explore the dependence of some of these impacts of urbanization on the dataset chosen for select use cases and discuss the importance of these uncertainties for both modeling and observational estimates of urban climate and environmental risks. Our results demonstrate the importance of choosing application-appropriate datasets for examining specific aspects of historical, present, and future urbanization with potential implications for informing sustainable development, resource allocation, and quantifying urban climate impacts.