No abstract

This paper presents the analysis of Earth Observation data records collected between 1975 and 2014 for assessing the extent and temporal evolution of the built-up surface in the frame of the Global Human Settlement Layer project. The scale of the information produced by the study enables the assessment of the whole continuum of human settlements from rural hamlets to megacities. The study applies enhanced processing methods as compared to the first production of the GHSL baseline data. The major improvements include the use of a more refined learning set on built-up areas derived from Sentinel-1 data which allowed testing the added-value of incremental learning in big data analytics. Herein, the new features of the GHSL built-up grids and the methods are described and compared with the previous ones using a reference set of building footprints for 277 areas of interest. The results show a gradual improvement in the accuracy measures with a gain of 3.6% in the balanced accuracy, between the first production of the GHSL baseline and the latest GHSL multitemporal built-up grids. A validation of the multitemporal component is also conducted at the global scale establishing the reliability of the built-up layer across time.

The Global Human Settlement Layer (GHSL) produces new global spatial information, evidence-based analytics describing the human presence on the planet that is based mainly on two quantitative factors: (i) the spatial distribution (density) of built-up structures and (ii) the spatial distribution (density) of resident people. Both of the factors are observed in the long-term temporal domain and per unit area, in order to support the analysis of the trends and indicators for monitoring the implementation of the 2030 Development Agenda and the related thematic agreements. The GHSL uses various input data, including global, multi-temporal archives of high-resolution satellite imagery, census data, and volunteered geographic information. In this paper, we present a global estimate for the Land Use Efficiency (LUE) indicator—SDG 11.3.1, for circa 10,000 urban centers, calculating the ratio of land consumption rate to population growth rate between 1990 and 2015. In addition, we analyze the characteristics of the GHSL information to demonstrate how the original frameworks of data (gridded GHSL data) and tools (GHSL tools suite), developed from Earth Observation and integrated with census information, could support Sustainable Development Goals monitoring. In particular, we demonstrate the potential of gridded, open and free, local yet globally consistent, multi-temporal data in filling the data gap for Sustainable Development Goal 11. The results of our research demonstrate that there is potential to raise SDG 11.3.1 from a Tier II classification (manifesting unavailability of data) to a Tier I, as GHSL provides a global baseline for the essential variables called by the SDG 11.3.1 metadata.

Abstract:In the last few decades the magnitude and impacts of planetary urban transformations have become increasingly evident to scientists and policymakers. The ability to understand these processes remained limited in terms of territorial scope and comparative capacity for a long time: data availability and harmonization were among the main constraints. Contemporary technological assets, such as remote sensing and machine learning, allow for analyzing global changes in the settlement process with unprecedented detail. The Global Human Settlement Layer (GHSL) project set out to produce detailed datasets to analyze and monitor the spatial footprint of human settlements and their population, which are key indicators for the global policy commitments of the 2030 Development Agenda. In the GHSL, Earth Observation plays a key role in the detection of built-up areas from the Landsat imagery upon which population distribution is modelled. The combination of remote sensing imagery and population modelling allows for generating globally consistent and detailed information about the spatial distribution of built-up areas and population. The GHSL data facilitate a multi-temporal analysis of human settlements with global coverage. The results presented in this article focus on the patterns of development of built-up areas, population and settlements. The article reports about the present status of global urbanization (2015) and its evolution since 1990 by applying to the GHSL the Degree of Urbanisation definition of the European Commission Directorate General for Regional and Urban Policy (DG-Regio) and the Statistical Office of the European Communities (EUROSTAT). The analysis portrays urbanization dynamics at a regional level and per country income classes to show disparities and inequalities. This study analyzes how the 6.1 billion urban dwellers are distributed worldwide. Results show the degree of global urbanization (which reached 85% in 2015), the more than 100 countries in which urbanization has increased between 1990 and 2015, and the tens of countries in which urbanization is today above the global average and where urbanization grows the fastest. The paper sheds light on the key role of urban areas for development, on the patterns of urban development across the regions of the world and on the role of a new generation of data to advance urbanization theory and reporting.

Spatially consistent and up-to-date maps of human settlements are crucial for addressing policies related to urbanization and sustainability, especially in the era of an increasingly urbanized world. The availability of open and free Sentinel-2 data of the Copernicus Earth Observation program offers a new opportunity for wall-to-wall mapping of human settlements at a global scale. This paper presents a deep-learning-based framework for a fully automated extraction of built-up areas at a spatial resolution of 10 m from a global composite of Sentinel-2 imagery. A multi-neuro modeling methodology building on a simple Convolution Neural Networks architecture for pixel-wise image classification of built-up areas is developed. The core features of the proposed model are the image patch of size 5 × 5 pixels adequate for describing built-up areas from Sentinel-2 imagery and the lightweight topology with a total number of 1,448,578 trainable parameters and 4 2D convolutional layers and 2 flattened layers. The deployment of the model on the global Sentinel-2 image composite provides the most detailed and complete map reporting about built-up areas for reference year 2018. The validation of the results with an independent reference dataset of building footprints covering 277 sites across the world establishes the reliability of the built-up layer produced by the proposed framework and the model robustness. The results of this study contribute to cutting-edge research in the field of automated built-up areas mapping from remote sensing data and establish a new reference layer for the analysis of the spatial distribution of human settlements across the rural–urban continuum.