According to the “2018 Revision of World Urbanization Prospects” (UN DESA), “in today’s increasingly global and interconnected world, 55% of the world’s population lives in urban areas. The coming decades will bring further profound changes to the size and spatial distribution of the global population, also due to increasing migration processes. The continuing urbanization is projected to add 2.5 billion of urban population by 2050, and the world population living in cities to increase to 68%.” Under these circumstances, sustainable development challenges will increasingly concentrate in cities.

Under the umbrella of the ERA-PLANET Network, the SMURBS project (H2020) brings together and exploits synergies between different EO platforms, namely in situ networks, satellites, state-of-the-art models, and new technologies (e.g. smart sensors), while also engaging citizens in collecting and sharing data. This integration paves the way for mainstreaming EO solutions into urban development tools and practices to enhance environmental and societal resilience to air quality, natural and manmade disasters, city growth, migration pressures and health impacts. The smart city approach that SMURBS brings forward, based on and empowered by EO, is presented in Figure 1.

The long-term sustainability of any solution necessitates the early and meaningful engagement of stakeholders. Solutions should be addressing real needs in a practical way; thus, SMURBS has undertaken a number of relevant activities to meet and discuss with decision makers and citizens. To pursue this: i) a targeted online survey collected the current status and user needs from 86 stakeholders from different groups (i.e. public authorities, academia, industry), ii) a stakeholder workshop with 26 participants resulted in specifications on actual needs, iii) 37 dedicated interviews with high-level decision makers gave valuable ideas for the co-design of solutions, iv) a citizen workshop gave emphasis on local issues and particularities, as well as citizen engagement to environmental issues and v) an industry workshop with SME experts on air quality sensors initiated discussion and interactions for technological solutions and interoperability aspects. The input received is well-distributed, both stakeholder- and geographical-wise. Public authorities and academia, from 18 countries and 32 cities, are the main contributors (80%), while the rest is shared between NGOs/citizen associations and the EO private sector.

Key messages for tackling air quality urban pressures include the need for higher spatial resolution, real-time and forecast information, systematic measurement of new pollutants, source apportionment and AQ-health impact delineation. In the field of disaster management, higher spatio-temporal resolution, enrichment of EO-information layers with supplementary data from multiple sources and scientifically guided crowd-sourcing, are highlighted. Pressures in the area of urban growth are mostly related to urban densification, urban sprawl, loss of green areas, soil sealing, also under the prism of delineating their links with natural disasters and micro-climate), while the need for smart indicators to support, both directly and indirectly, the SDG frame are put forward. Finally, horizontal challenges are identified that include the EO incorporation into authorities’ operational procedures, the long term continuity of services, the collaboration between science and decision makers, and the empowerment of citizen’s participation.

SMURBS portfolio of smart urban solutions is currently composed of 21 solutions for air quality and health impacts, 5 solutions for disasters and 8 solutions for urban growth including migration, all tailored according to the identified user needs, and developed in a co-design and co-creation fashion with stakeholders, in a dynamic way. The majority of solutions uptake, in multiple ways, Copernicus services such as CAMS, CLMS and EMS, while validation exercises are also foreseen within the project.

Indicative examples of solution types per theme include: (Air Quality) methodologies to integrate low-and mid-cost AQ sensors with specialized measurements from urban super sites, regional to city scaling modeling for scenario based assessments and forecasting, real time source apportionment techniques, mobile applications for forecasts, past data analytics, crowdsourcing of air quality perception by citizens; (Disasters) incorporation of crowdsourcing into pre- real time and post-assessment of floods, 2D/3D mapping of critical urban infrastructure, peat fire detection services, land deformation mapping; (Urban Growth) VHRL for urban change detection, Copernicus-compliant Urban Atlas and validation, disaster risk assessment for refugee areas and other resilience indicators.

These solutions will be tested and applied in more than 20 pilot cities, not limited to Europe but also in other continents, in alignment with ERA-PLANET goal to o strengthen the European Research Area in the domain of Earth Observation in coherence with the European participation to Group on Earth Observation (GEO) and Copernicus. SMURBS brings forward EO as a valuable and irreplaceable means to increase “smartness” of future cities and support informed decision making. In close collaboration with EO4SDG, GEO-Essential and National Statistics Offices (NSOs), the project promotes EO integration with other source of information into effective urban data analytics tools, to support the monitoring and achievement of the Sustainable Development Goals (SDG). In this direction, SMURBS is working together with the rest three ERA-PLANET projects to develop specific workflows, from Essential variables to monitoring of SDG indicators and SDG11 in particular, to be included in the ERA-PLANET Virtual Lab.

1.0 Introduction
During the UN Habitat III in Quito (2016), countries agreed to implement National Urban Policies (NUP) and plans which are required for sustainable Cities. The UN Habitat defines a NUP as ..”..A coherent set of decisions derived through a deliberate government-led process of coordinating and rallying various actors for a common vision and goal that will promote more transformative, productive, inclusive and resilient urban development for the long term (UN-Habitat, 2016).” The NUPs are also needed for implementing the UN Sustainable Development Goal 11 “making cities inclusive, safe, resilient and sustainable.” Urban planning based on quantitative data is increasingly important in developing countries where urban population pressures result in urban sprawl and development of informal settlements which lack basic services for the residents. Thus geo-spatial data play a major role in evidence based urban planning. However, one of the main challenges that City Authorities face in developing countries is the lack of high quality validated geo-spatial data, as well as the technical capacities to use the data for planning. In this context the European Space Agency (ESA) has been collaborating with Multi-Lateral Development Banks (MDBs) and counterparts in developing countries to provide technical support for the improved understanding and utilisation of Earth Observation (EO) satellite data and the corresponding geo-spatial information. This paper will present the outcome of the ESA EO4SD Urban project’s support to MDB programmes involved in implementing aspects of the New Urban Agenda from the UN Habitat III.

2.0 EO4SD Urban support to the new Urban Agenda
In 2016 a Consortium of European EO Service Providers led by GAF AG (Germany) started a 3 year project to support pre-operational and operational MDB Urban Development programmes with a suite of geo-spatial data. The project has an overall objective to introduce the advancements of EO technologies to the Banks, as well as to support the mainstreaming of EO data into the work practices of Municipalities/City Planners in developing countries. This objective is being achieved via the application of geo-spatial products for various urban development programmes. In Phase 1 of the project, 16 global Cities serving 3 MDBs were mapped and for each City, Baseline Land Use/Land Cover (LU/LC) data for two points in time (between 2015/2016 and 2005/2006) as well as some derived products such as Green Areas, Transport Networks, Informal Settlements, Population Density were provided to the Banks and their City counterparts. Some Special products, such as Building Heights, Terrain Motion and Flood Hazard/Risk were also provided. In total, approximately 204 products and 132 maps were delivered. For Phase 2 another 15 Cities will be mapped. For core urban areas Very High Resolution (VHR) EO data such as Quickbird and Pleiades were used with a minimum mapping unit (MMU) of 0.25ha and Sentinel and Landsat were used for mapping the peri-Urban regions. The overall accuracies achieved for the various products ranged from 85-95% which can be considered as very accurate.

3.0 Geo-Spatial Products for Quantitative Assessments and Diagnostics
Several projects at World Bank (WB), Inter-American Development Bank (IADB) and the Asian Development Bank (ADB) were supported with the EO4SD Urban products. In all cases the products contributed to better understanding of the spatial relation of land use types and different settlement patterns, which are required by Cities for urban planning, zoning and monitoring urban growth; these aspects are part of a Cities efforts at preparing NUPs.

A direct example of the utility of the geo-spatial products for urban policy development and planning was the application in the WB Technical Assistance project “Urban Planning Study-Impact & Effectiveness of Urban Planning on City Spatial Development” for Tanzania. The Tanzania Strategic Cities Project is supporting selected local government authorities, in the preparation and enforcement of General Planning Schemes or Master Plans, detailed area plans and urban design plans. The EO4SD Urban project provided the baseline Land Use/Land Cover products for three Secondary Cities of Arusha, Dodoma and Kigoma; the products were used to assess existing Master Plans and analyze urban growth spatially over almost a decade in these 3 Cities. See Figure 3 for an example of a land use change map for Dodoma (depicting densification). Additional products such as Road Networks, Green Areas and Informal Settlements were also provided which were used in the spatial analytic work. The Urban Extent product produced by the German Space Agency provided a basis for the assessment of spatial metrics such as fragmentation and leapfrog growth in the Cities. Results of the study provide Cities with detailed spatial information on location of urban growth, especially the development of unplanned settlements which require then additional infrastructure development (WB, 2018). The granularity also provided by the geo-spatial data provide the City Authorities with information on land use practices, densification and where redevelopment has to be planned.

Another case relates to the application of the land use and derived products for improved land use planning in 2 Indonesian Cities of Denpasar and Semarang which are part of the WB City Planning Labs (CPL) programme. This programme has the overall objective of “establishing technical capacity at the Municipal level to provide reliable analytical support to a city’s planning, policy and infrastructure decisions…” (WB, 2016). The city of Semarang and Denpasar partnered with the CPL Programme to develop capacity, undertake geospatial planning and data sharing. Both Cities were interested to use remote sensing to develop spatial data related to urban growth, Land Use/Land Cover changes, transportation infrastructure and accessibility, population density/distribution, and land subsidence and flooding issues in both cities. The updated spatial information from the EO4SD project could be used for diagnostics for the improvement of the urban planning in the Cities.

The ADB is funding the City of Kolkata via the “Kolkata Environmental Improvement Investment Program” which has a main focus on improving land-use management and sustainable urban planning. The main stakeholder was the Kolkata Municipal Corporation (KMC) and the geo-spatial products provided specifically supported KMC with the possibility for infrastructure planning such as water pipelines for informal settlements and the land-cover and land-use maps assisted with the identification of areas prone to disasters hence strengthening climate resilience. The KMC and ADB further noted that the products were being used for designing early flood forecasting and warning system.

6. Conclusions
The three projects presented in this Abstract demonstrate the high utility of remote sensing and geo-spatial products as inputs for evidence based approaches to urban planning which contribute to NUPs. All the cases in Phase 1 of the project, demonstrated high levels of product utility for urban planning, and positive feedback from the different Bank projects and clients was received. It is anticipated that at the end of the EO4SD Urban project in 2019 there will be the possibility to showcase for 32 global cities and MDB programmes the value of remote sensing for improved decision making in urban planning and policy developments; this should further spur the application of the technology in a regular and operational manner in MDB programmes and in developing countries.

References:
UN Habitat, 2016, “National Urban Policy: A Guiding Framework,” First published in Nairobi in 2015 by UN-Habitat, Revised in 2016
World Bank, 2016, “Kota Semarang, Urban Growth and Access to Public Infrastructure, CPL Semarang Report,” Final Draft
World Bank, 2018, “Urban Planning Study for Tanzania –Impact and Effectiveness of Urban Planning on City Spatial Development,” © World Bank

Session: D.2.04 EO for Resilient Cities


Keywords: Disaster Risk Reduction in Cities, City resilience


Understanding disaster risk and resilience to climate hazards starts with measuring their exposure to hazards. This work provides a global assessment of the exposure of all world cities over 50000 inhabitants to the impact of four hazards: earthquakes, floods, sea level surge and heat waves. It also produces the land use efficiency indicator by combining population density and built up over the extent of the city.
Cities are the center of human activities as most people live and work in settlements. Settlements also host what is most valuable to societies. Protecting cities and their inhabitants from the impact of natural hazard is a priority for city managers and policy makers. At the international level disaster risk reduction and climate resilience is one of key priorities for funding agencies. In fact, the Sendai Framework for Disaster Risk Reduction, The Paris Climate Agreements, and the Sustainable Development Goals, all include indicators that address hazard impact and resilience.
The analysis provides historical trends of the change in exposure over time by intersecting hazard layers with a database of over 12000 cities. The analysis uses two global human settlement layers, that of global built-up area extracted by processing 40 years of Landsat images, and that of population density produced by combining built-up and population census. In addition, the two data layers available for 2015 are then used to define the spatial extent of cities and settlements based on The New Degree of Urbanization. The New Degree of Urbanization is a model that uses population density, built-up and spatial adjacency criteria to generate a standardized global information layer of cities and settlements. This work shows only the over 12000 cities with population over 50000 inhabitants. The New Degree of Urbanization work is part of an international cooperation between European Commission, Organization for Economic Cooperation and Development, the Food and Agricultural Organization, the United Nations Statistical Division, the Organization for Economic Cooperation and Development, and The World Bank that aims to generate a standardized definition of cities and settlements worldwide independent from national criteria’s.
The results of the analysis shows four global maps of cities exposed to hazards, with associated database. The data are available as open source data. For each hazard we also provides the trends in built-up and population exposure and thus provide insights into the countries and continents that are mostly affected by hazards as well as a listing of all the cities that are exposed to more than one hazard. The extensive list of cities and data will be available as an open source data sets available for further analysis. The baseline data produced from Earth observation provide also the building blocks from which to start a more holistic assessment of city resilience.

Reliably monitoring global urbanization is of key importance to accurately estimate the distribution of the continually expanding population, along with its effects on the use of resources (e.g., soil, energy, water), infrastructure and transport needs, socioeconomic development, human health, food security, etc. To this purpose, since the last decade several global maps outlining settlements have started being produced by means of satellite imagery. In this framework, the two currently most largely employed are JRC’s Global Human Settlement Layer (GHSL) derived at 30m spatial resolution from Landsat optical data and, especially, DLR’s Global Urban footprint (GUF) derived at 12m spatial resolution from TanDEM-X/TerraSAR-X radar data. However, it is worth noting that, despite generally accurate, these layers still exhibit both some over- and underestimation issues. Specifically, this is mostly due to the fact that they have been generated by means of: i) single-date scenes (which can be strongly affected by the specific acquisition conditions) and ii) solely using either optical or radar data, which are sensible to different structures on the ground (e.g., with optical imagery bare soil and sand generally tend to be misclassified as settlements, while this does not occur with radar data; on the contrary, with radar imagery complex topography areas or forested regions can be wrongly categorized as settlements, whereas this does generally not happen if optical data are employed).
In order to overcome these limitations, we have developed a novel methodology that jointly exploits for the first time mass multitemporal optical and radar data. In particular, the rationale is that the temporal dynamics of human settlements over time are sensibly different than those of all other information classes. Hence, given all the multitemporal images available over a region of interest in the selected time interval, we first extract key temporal statistics (i.e., temporal mean, minimum, maximum, etc.) of: i) the original backscattering value in the case of radar data; and ii) different spectral indices (e.g., vegetation index, built-up index, etc.) derived after performing cloud masking in the case of optical imagery. Then, different classification schemes based on advanced machine learning are separately applied to the optical and radar temporal features, respectively, and, finally, the two outputs are properly combined together.
In the light of its great robustness, the method has been recently used for generating the new World Settlement Footprint (WSF) 2015, i.e. a 10m resolution binary mask outlining the extent of human settlements globally derived by means of 2014-2015 multitemporal Landsat-8 and Sentinel-1 imagery (of which ~420,000 and ~250,000 scenes have been processed, respectively). Furthermore, to quantitatively assess the high accuracy and reliability of the layer we have carried out in collaboration with Google an unprecedented validation exercise based on a huge amount of ground-truth samples (i.e. 900,000) labelled by crow-sourcing photointerpretation. In particular, a statistically robust and transparent protocol has been defined following the state-of-the-art practices currently recommended in the literature. Overall, results assess the great effectiveness of the WSF2015, which also outperforms all other currently existing similar global layers including the GUF and the GHSL.
Nevertheless, to properly analyze and understand the complexity of human settlements and ensure their sustainable development, not only information about the current extent is sufficient. Rather, outlining their growth over time is also fundamental for modelling ongoing trends and implementing dedicated suitable planning strategies. However, so far the existing products are mostly available for few time steps in the past and their quality– yet by simple qualitative assessment against e.g. Google Earth historical imagery – appears rather poor.
To address this issue, we designed and implemented a novel iterative technique for outlining the past settlement extent from Landsat multitemporal imagery available from late 1984 (indeed no comparable radar dataset is continuously and globally consistently available over the last 30 years). First, under the assumption that pixels categorized as non-settlement at a later time cannot be marked as settlement at an earlier time, all areas excluded from the WSF2015 are discarded a priori from the analysis. Next, for each target year in the past all available Landsat scenes acquired over the investigated area of interest are gathered and cloud masking is performed. Key temporal statistics (i.e., temporal mean, minimum, maximum, etc.) are then extracted for different spectral indices including the normalized difference built-up index (NDBI), the normalized difference vegetation index (NDVI) and the modified normalized difference water index (MNDWI). Going backwards in time, training samples for the given target year are iteratively extracted by applying morphological filtering to the settlement mask derived for the previous time step, as well as excluding potentially mislabeled samples by adaptive thresholding on the temporal mean NDBI, MNDWI and NDVI. Finally, random forest classification in performed.
Extensive experimental analyses over several test sites assessed the great effectiveness and robustness of the methodology. Accordingly, it is currently being employed within the Google Earth Engine environment for generating the WSF Evolution, i.e. a novel dataset aimed at outlining the growth of settlement extent globally at 30m spatial resolution and high temporal resolution (i.e., 5-year or even finer) from 1985 to 2015. The WSF Evolution is envisaged to be completed in the next few months and it is expected to become a revolutionary product in support to a variety of end users in the framework of several thematic applications, helping to understand, as never before, how settlements evolved over three decades while capturing specific temporal trends.

Urban population has grown rapidly in recent years and it is expected that over 1.4 billion people will move into cities by 2030. The 2030 Agenda for the Sustainable Development Goals developed by the United Nations gives a prominent role to monitoring and studying the urbanization process. Satellite data are playing a fundamental role to retrieve the urban footprint at regional and global scales. Several methodologies have been developed using multispectral and/or Synthetic Aperture Radar (SAR) imagery and relevant research projects have been carried out to generate global datasets as the Global Urban Footprint (GUF) by the German Aerospace Center (DLR) or the Global Human Settlement Layer (GHSL) by the Joint Research Center (JRC) [1] [2] . In particular, these datasets provide a reliable and global map of urban areas, but they are characterized by low temporal resolution (i.e. every five years) which highlights the need of further research and method development [3].

The aim of this work is to develop a globally applicable and entirely automatic method to continuously monitor urban footprints using Sentinel-1 and Sentinel-2 dense time series exploiting the Google Earth Engine (GEE) cloud platform [4]. A Free and Open Source GEE App is under development to enable users to extract the Urban Footprint of their selected area and the corresponding sensing period of interest. Through this easy and ready to use tool users will be able to analyse urban areas, allowing comparison with other existing methods and data. The innovative aspect of the developed method is to combine Sentinel-1 and Sentinel-2 dense time series using a totally unsupervised approach. The estimation of the selected urban footprint is performed in several progressive steps. First, the area of interest is divided into mountainous and non-mountainous areas using an available DSM (i.e. SRTM or ALOS World 3D) to take into account the layover and foreshortening SAR geometric distortions. Then, Sentinel-1 ascending and descending time series are processed in order to enhance the backscatter of stable urban areas and to compute the Sentinel-1 Urban mask using an automatic thresholding procedure. The latest step is to compute a probability urban map combining the Sentinel-1 Urban mask with several spectral indexes (i.e. NDVI, NDBI, NDWI) generated from cloud-free Sentinel-2 time series if available or a cloud-free image composite, computed using all available Sentinel-2 images, acquired during the selected sensing period.

We applied the developed approach in several cities worldwide (i.e. Beijing, Lagos, Milan, Mumbai, New York, Rio and Stockholm) in order to evaluate its potentialities in areas characterized by different urban density and morphologies. We computed the urban footprint in different periods to evaluate the temporal stability of the method and to produce urban footprint time series. The results show that through this method it is possible to obtain high accuracy (kappa higher than 85 percent) with respect to the reference data acquired within the EO4Urban project in all cities and in different periods. The developed method obtains equal or higher accuracy than GUF and GHSL data in the same area, and a visual comparison shows that the integration of the Sentinel optical and SAR data leads to achieving highly detailed information.

[1] T. Esch et al., “Urban footprint processor-Fully automated processing chain generating settlement masks from global data of the TanDEM-X mission,” IEEE Geosci. Remote Sens. Lett., vol. 10, no. 6, pp. 1617–1621, 2013.
[2] M. Pesaresi et al., “The global human settlement layer from landsat imagery,” 2016 IEEE Int. Geosci. Remote Sens. Symp., pp. 7276–7279, 2016.
[3] T. Esch et al., “Where we live-A summary of the achievements and planned evolution of the global urban footprint,” Remote Sens., vol. 10, no. 6, 2018.
[4] N. Gorelick, M. Hancher, M. Dixon, S. Ilyushchenko, D. Thau, and R. Moore, “Google Earth Engine: Planetary-scale geospatial analysis for everyone,” Remote Sens. Environ., vol. 202, pp. 18–27, 2017.

Heatwaves are a serious threat to the health and wellbeing of the continuously growing urban population and are projected to worsen under climate change. The European Environmental Agency (EEA) in its 2017 report “Climate change, impacts and vulnerability in Europe” stresses that extreme temperature events affect all Europe and cause many excess deaths. To that end future projections suggest that over the next decades heatwaves will occur more often in Europe and that there will be a substantial increase in heat-related mortality. In detail, it is estimated that without adaptation and physiological acclimatization, heat-related mortality in Europe will increase by 60,000 to 165,000 deaths per year by the 2080s (compared with the present baseline), with the highest impacts in southern Europe. To address this problem and ensure the safety and well-being of the urban population, well-informed disaster risk reduction (DRR) actions, where science and technology play a key role, are required. EXTREMA project (EXTReme tEMperature Alerts for Europe; https://extrema.space/), a recently-funded civil protection project from European Commission’s Directorate-General for European Civil Protection and Humanitarian Aid Operations (DG-ECHO), uses satellite thermal data retrieved in real-time as well as model output so as to provide actionable knowledge (i.e. knowledge that is useful, usable and used) to citizens and city authorities. This information has the form of personalized heatwave risk estimations that can be readily used by the general public though the EXTREMA mobile application and by the city authorities through the EXTREMA dashboard. The EXTREMA mobile application in particular aims to inform the user on his/hers personal extreme temperature health risk, when the temperature reaches or exceeds certain levels based on their profile. To achieve the above, the mobile app requires spatially distributed, timely and accurate air temperature (TA) information. In the context of EXTREMA this information is retrieved from 5-min thermal image data acquired by the Spinning Enhanced, Visible and Near-Infrared Imager (SEVIRI) by assimilating them with Numerical Weather Predictions (NWP) from the Global Forecasting System (GFS) (Keramitsoglou, et al., 2016). The output TA data, which are also gap-filled and downscaled to a finer spatial resolution, can then be accessed by the mobile application and interpreted as risk estimation. In particular, the mobile application using the geolocation functionality of the smartphone can identify the location of the user and then retrieve the corresponding satellite-derived TA from a dedicated web-service. This TA is then compared to a set of predefined air temperature thresholds (derived from the published epidemiological literature), which are linked to heatwave risk levels (no risk; low risk; increased risk; and high risk). In the cases where the personalized risk is increased or high, the mobile app also offers recommendation on what to do based on the World Health Organization guidelines and routing directions for the nearest cooling centre, e.g. air-conditioned public or private spaces, community centres, parks, fountains, or hospitals if necessary. The app was launched in summer 2018 in Paris, Rotterdam and Athens, three members of the 100 Resilient Cities international network, pioneered by Rockefeller Foundation.


References:
Keramitsoglou, I., C. Kiranoudis, P. Sismanidis, and K. Zakšek (2016), An Online System for Nowcasting Satellite Derived Temperatures for Urban Areas, Remote Sens., 8(4), 306, doi:10.3390/rs8040306.
Keramitsoglou, I. et al. (2017), Urban thermal risk reduction: Developing and implementing spatially explicit services for resilient cities, Sustain. Cities Soc., 34, 56–68, doi:10.1016/j.scs.2017.06.006.