WSF Product Suite GUF 2012 World Settlement Footprint - WSF ▪ Innovation ▪ Free and open input data (Sentinel-1, Sentinel-2, Landsat) ▪ Improved spatial and thematic accuracy (multi-sensor, multi-date) WSF 2015 ▪ Full update capability (monthly – yearly) ▪ Extended product range World Settlement Footprint 2015 Binary mask outlining the extent of human settlements globally for the year 2015. Spatial Resolution 10m x 10m Spatial Coverage Global Publication Marconcini et al., 2020 Availability Open and free Access DLR EOC Geoservice (https://geoservice.dlr.de/web) Exemplary Use Cases • Detecting urban areas in natural hazard risk zones, • Determining built-up density • Determining distribution of and access to hospitals, schools etc. in urban areas WSF 2015 World Settlement Footprint 2019 Binary mask outlining the extent of human settlements globally for the year 2019. Spatial Resolution 10m x 10m Spatial Coverage Global Publication Marconcini et al., 2021 Availability Open and free Access DLR EOC Geoservice (https://geoservice.dlr.de/web) Exemplary Use Cases • Detecting urban areas in natural hazard risk zones, • Determining built-up density • Determining distribution of and access to hospitals, schools etc. in urban areas WSF 2015 WSF 2019 final version World Settlement Footprint Evolution Binary mask outlining the extent of human settlements on a yearly basis from 1985 to 2015. Spatial Resolution 30m x 30m Spatial Coverage Global Publication Marconcini et al., 2021 Availability Open and free Access DLR EOC Geoservice (https://geoservice.dlr.de/web) Exemplary Use Cases • Tracking development of urban areas in zones prone to natural hazards • Developing long term trends on urban growth • Analysing urban growth and urban growth mechanisms (infill, expansion and leapfrog) First time monitoring of global urbanization WSF evolution 1985-2015 over three decades Processing of entire Landsat-5/7 archive (~7,000,000 scenes) on a yearly basis Sacramento The dataset outlines the growth of settlement extent globally at 30m spatial resolution on a yearly basis from 1985 to 2015; The layer has been generated with the support of the Google Earth Engine team by processing the whole Landsat-5/7 archive including ~7,000,000 scenes; Accuracy assessment based on ~1,200,000 validation samples collected by crowd-sourcing completed. WSF evolution 1985-2015 Sacramento WSF evolution 1985-2015 Dar Es Salaam WSF evolution 1985-2015 Dar Es Salaam WSF evolution 1985-2015 World Settlement Footprint Imperviousness The WSF Imperviousness dataset quantifies the percent of impervious surface within each 10mx10m settlement pixels reported by the WSF2019. Spatial Resolution 10m x 10m Spatial Coverage Global Publication Pending Publication Availability Planned release in Autumn 2022 (Q3). Access Pending Exemplary Use Cases • Population estimation and distribution • Distribution of income levels • Determination of built-up density WSF 2019 imperviousness Cape Town WSF2019 imperviousness • Percent impervious surface (PIS) is generally a good proxy for the building density and is of key support for better distributing population in urban models, assessing the risk of urban floods or characterizing the urban heat island phenomenon; • Accuracy assessment ongoing. WSF 2019 imperviousness PERCENT IMPERVIOUS SURFACE World Settlement Footprint Population The WSF 2019 Population quantifies the number of people per pixel. It has been created using population estimates and vector data collected and harmonized by CIESIN for the year 2019. Spatial Resolution 10m x 10m Spatial Coverage Global Publication Palacios-Lopez et al., 2020 Availability Planned release in Autumn 2022 (Q3). Access Pending Exemplary Use Cases • Identification of number people affected by natural hazards • Estimating number of people per building Cape Town WSF2019 population • Estimating population density is fundamental for implementing efficient government policies and, in turn, allocating financial resources, planning intervention and quantifying people at risk; • Idea: exploiting the WSF 2019 imperviousness as proxy for distributing people in a target study region given their total amount (e.g., from census data, UN/World Bank estimates). INHABITANTS PER PIXEL (10x10m) World Settlement Footprint 3D The WSF3D quantifies the fraction (BF), total area (BA), average height (BH) and total volume (BV) of buildings for a measuring grid of 90m cell size. Spatial Resolution 90m x 90m Spatial Coverage Global Publication Esch et al., 2022 Availability Waiting for permission granted by the German ministry of defence for open and free release. Access Pending Exemplary Use Cases • Estimation of real-estate values in data poor environments • Estimating number of inhabitants per building Tokio WSF 3D 90m WSF 3D 90m Building fraction Building volume New York WSF 3D 90m WSF 3D 90m Building fraction Building volume Fraction Volume 0 100 Low High [%] 3 [m ] WSF 3D Cairo WSF 3D References • Marconcini, M., Metz-Marconcini, A., Üreyen, S., Palacios-Lopez, D., Hanke, W., Bachofer, F., ... & Strano, E. (2020). Outlining where humans live, the World Settlement Footprint 2015. Scientific Data, 7(1), 1-14. • Marconcini, M., Metz-Marconcini, A., Esch, T., & Gorelick, N. (2021). Understanding current trends in global urbanisation-the world settlement footprint suite. GI_Forum, 9(1), 33-38. • Palacios-Lopez, D., Bachofer, F., Esch, T., Marconcini, M., MacManus, K., Sorichetta, A., ... & Reinartz, P. (2021). High-resolution gridded population datasets: Exploring the Capabilities of the world settlement footprint 2019 imperviousness layer for the African continent. Remote Sensing, 13(6), 1142. • Esch, T., Brzoska, E., Dech, S., Leutner, B., Palacios-Lopez, D., Metz-Marconcini, A., ... & Zeidler, J. (2022). World Settlement Footprint 3D-A first three-dimensional survey of the global building stock. Remote Sensing of Environment, 270, 112877. Use Case Collection Population Distribution in Africa: Disaggregation of Population Statistics using WSF 2019 Imperviousness Reliability for total of 52 countries: 35 = high (< 25 % MAE) 16 = medium (25-50 % MAE) 1 = poor (>50 % MAE) Palacios-Lopez, D., Bachofer, F., Esch, T., Marconcini, M., MacManus, K., Sorichetta, A., Zeidler, J., Dech, S., Tatem, A.J., Reinartz, P. (2021): High-resolution gridded population datasets: Exploring the capabilities of the World Settlement Footprint 2019 Imperviousness layer for the African continent. Remote Sens. 2021. Urbanization Africa: Identification of urban centers in Africa (> 10,000 inhabitants) Exposure to Natural Hazards • Detection of urban clusters with > 10,000 inhabitants • Number of population exposed to natural hazards per urban cluster • Floods, Earthquakes, Landslides at four hazard levels Population exposed to landslide hazard at Hazard Level 1 Urbanization Africa: Urban Growth Mechanisms Calculated for 87 prioritized cities in Africa between the year 2000 and 2015 Extension: Built-up area added in the new period and constituting contiguous urban clusters that are attached to the urban extent of the earlier period. Infill: Built-up area added in the new period that developed in open space within the urban extent of the earlier period. Leapfrog: Built-up area added in the new period and constituting contiguous urban clusters that are not attached to the urban extent of the earlier period. Is Africa really facing urban sprawl? Northern Africa Western Africa Middle Africa Eastern Africa Southern Africa Size of the circles proportional to log(total population) Potential hazard zones in Matola, Mozambique This model shows the number of inhabitants per individual building mapped, combined in this image with potential hazard zones. The red areas indicate high vulnerability for Matola, Mozambique. Landslide danger In this image, landslide danger is combined with population distribution for the whole of Africa. The red colours indicate regions of high risk. WSF Population: Mapping of major non-residential areas by joint analyzing WSF layers Major Non-residential areas (e.g. airports, ports, stadiums, factories, large-commercial centers) can be distinguished from other build-up structures due to their high values in all relevant WSF-layers (WSF3D, WSF Imperviousness). Masking off these areas can improve the modelling of population distribution at local scales. Munich area with major non-residential areas and their corresponding impervious and volume values. WSF Population: Mapping of major non-residential areas by joint analyzing WSF layers Machine learning approach to classify WSF built-up pixels into two main classes: 1. Residential: Include small mix-use buildings or small non-residential buildings like churches, gyms, schools. 2. Non-Residential: Major industrial areas, airports, ports and large commercial-centers Ground Truth Classification 85% 88% Residential Residential 15% 12% Non-Residential Non-Residential Residential (Percentage of built-up settlements) Non-Residential Monitoring of Urban Heat Island Effects › Local surface temperatures in urban Imperviousness environments are significantly higher than in their rural surroundings › This surface urban heat island effect is strongly driven by the amount and the distribution of artificial land cover › The dependency of the surface temperature on the imperviousness patterns can be quantified and serve as a measure of the susceptibility of an urban environment to climatic stress Land Surface Temperature Deployment of prototypic processor at Data and Information Access Services (DIAS) Pilot 3 - EO-based air pollution health risks profiling in urban environment WSF2015 Population Density Sentinel-5P Satellite-based NO2-Trend Global Service Platforms (e.g., U-TEP) 3 4 5 6 7 8 2007 2009 2011 2013 2015 2017 2019 NO2 [1015 molec/cm²] Approaches Comparison Polyphemus (high resolution model) CAMS (low resolution model) SO2 NO2 O3 In collaboration with DFD-ATM: Monitoring stations WSF 2015 – Population Density for Germany T. Erbertseder, M. Houdayer, L. Gilardi Outreach